Add comment

- eliminate reader function generation
- fix write to closed channel panic

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

@@ -2,15 +2,17 @@
 name: Bug/Issue report
 about: Create a report to help us improve
 title: ''
-labels: ''
+labels: ['triage-needed']
 assignees: ''
 
 ---
 
 **Describe the problem**
+
 A clear and concise description of what the problem is.
 
 **To Reproduce**
+
 Steps to reproduce the behavior:
 1. Go to '...'
 2. Click on '....'
@@ -18,13 +20,25 @@ Steps to reproduce the behavior:
 4. See error
 
 **Expected behavior**
+
 A clear and concise description of what you expected to happen.
 
+**Are you using NetBird Cloud?**
+
+Please specify whether you use NetBird Cloud or self-host NetBird's control plane.
+
+**NetBird version**
+
+`netbird version`
+
 **NetBird status -d output:**
-If applicable, add the output of the `netbird status -d` command
+
+If applicable, add the `netbird status -d' command output.
 
 **Screenshots**
+
 If applicable, add screenshots to help explain your problem.
 
 **Additional context**
+
 Add any other context about the problem here.

.github/ISSUE_TEMPLATE/feature_request.md

@@ -2,7 +2,7 @@
 name: Feature request
 about: Suggest an idea for this project
 title: ''
-labels: ''
+labels: ['feature-request']
 assignees: ''
 
 ---

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

@@ -32,8 +32,14 @@ jobs:
           restore-keys: |
             macos-go-
 
+      - name: Install libpcap
+        run: brew install libpcap
+
       - name: Install modules
         run: go mod tidy
 
+      - name: check git status
+        run: git --no-pager diff --exit-code
+
       - name: Test
-        run: NETBIRD_STORE_ENGINE=${{ matrix.store }} go test -exec 'sudo --preserve-env=CI' -timeout 5m -p 1 ./...
+        run: NETBIRD_STORE_ENGINE=${{ matrix.store }} go test -exec 'sudo --preserve-env=CI,NETBIRD_STORE_ENGINE' -timeout 5m -p 1 ./...

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

@@ -14,8 +14,8 @@ jobs:
   test:
     strategy:
       matrix:
-        arch: ['386','amd64']
-        store: ['jsonfile', 'sqlite']
+        arch: [ '386','amd64' ]
+        store: [ 'jsonfile', 'sqlite', 'postgres']
     runs-on: ubuntu-latest
     steps:
       - name: Install Go
@@ -36,13 +36,20 @@ jobs:
         uses: actions/checkout@v3
 
       - 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
+        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 }} go test -exec 'sudo --preserve-env=CI' -timeout 5m -p 1 ./...
+        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 ./...
 
   test_client_on_docker:
     runs-on: ubuntu-20.04
@@ -64,11 +71,14 @@ jobs:
         uses: actions/checkout@v3
 
       - 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
+        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 modules
         run: go mod tidy
 
+      - 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/
 
@@ -76,7 +86,7 @@ jobs:
         run: CGO_ENABLED=0 go test -c -o sharedsock-testing.bin ./sharedsock
 
       - name: Generate RouteManager Test bin
-        run: CGO_ENABLED=0 go test -c -o routemanager-testing.bin ./client/internal/routemanager/...
+        run: CGO_ENABLED=1 go test -c -o routemanager-testing.bin -tags netgo -ldflags '-w -extldflags "-static -ldbus-1 -lpcap"' ./client/internal/routemanager/...
 
       - name: Generate nftables Manager Test bin
         run: CGO_ENABLED=0 go test -c -o nftablesmanager-testing.bin ./client/firewall/nftables/...
@@ -103,7 +113,7 @@ jobs:
 
       - name: Run Engine tests in docker with file store
         run: docker run -t --cap-add=NET_ADMIN --privileged --rm -v $PWD:/ci -w /ci/client/internal -e NETBIRD_STORE_ENGINE="jsonfile" --entrypoint /busybox/sh gcr.io/distroless/base:debug -c /ci/engine-testing.bin  -test.timeout 5m -test.parallel 1
-      
+
       - name: Run Engine tests in docker with sqlite store
         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
 

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

@@ -46,7 +46,7 @@ jobs:
       - 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
 
       - 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 5m -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 -timeout 10m -p 1 ./... > test-out.txt 2>&1"
       - name: test output
         if: ${{ always() }}
         run: Get-Content test-out.txt

.github/workflows/golangci-lint.yml

@@ -19,7 +19,7 @@ jobs:
       - name: codespell
         uses: codespell-project/actions-codespell@v2
         with:
-          ignore_words_list: erro,clienta
+          ignore_words_list: erro,clienta,hastable,
           skip: go.mod,go.sum
           only_warn: 1
   golangci:
@@ -33,6 +33,10 @@ jobs:
     steps:
       - name: Checkout code
         uses: actions/checkout@v3
+      - 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
         with:
@@ -40,7 +44,7 @@ jobs:
           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
+        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
         with:

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

@@ -11,7 +11,7 @@ concurrency:
   cancel-in-progress: true
 
 jobs:
-  andrloid_build:
+  android_build:
     runs-on: ubuntu-latest
     steps:
       - name: Checkout repository
@@ -38,10 +38,10 @@ jobs:
       - 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 nebtird lib
+      - name: build android netbird lib
         run: PATH=$PATH:$(go env GOPATH) gomobile bind -o $GITHUB_WORKSPACE/netbird.aar -javapkg=io.netbird.gomobile -ldflags="-X golang.zx2c4.com/wireguard/ipc.socketDirectory=/data/data/io.netbird.client/cache/wireguard -X github.com/netbirdio/netbird/version.version=buildtest"  $GITHUB_WORKSPACE/client/android
         env:
           CGO_ENABLED: 0
@@ -56,10 +56,10 @@ jobs:
         with:
           go-version: "1.21.x"
       - name: install gomobile
-        run: go install golang.org/x/mobile/cmd/gomobile@v0.0.0-20230531173138-3c911d8e3eda
+        run: go install golang.org/x/mobile/cmd/gomobile@v0.0.0-20240404231514-09dbf07665ed
       - name: gomobile init
         run: gomobile init
-      - name: build iOS nebtird lib
-        run: PATH=$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 $GITHUB_WORKSPACE/client/ios/NetBirdSDK
+      - name: build iOS netbird lib
+        run: PATH=$PATH:$(go env GOPATH) gomobile bind -target=ios -bundleid=io.netbird.framework -ldflags="-X github.com/netbirdio/netbird/version.version=buildtest" -o ./NetBirdSDK.xcframework ./client/ios/NetBirdSDK
         env:
           CGO_ENABLED: 0

.github/workflows/release.yml

@@ -7,17 +7,7 @@ on:
     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"
@@ -106,6 +96,27 @@ jobs:
           name: release
           path: dist/
           retention-days: 3
+      -
+        name: upload linux packages
+        uses: actions/upload-artifact@v3
+        with:
+          name: linux-packages
+          path: dist/netbird_linux**
+          retention-days: 3
+      -
+        name: upload windows packages
+        uses: actions/upload-artifact@v3
+        with:
+          name: windows-packages
+          path: dist/netbird_windows**
+          retention-days: 3
+      -
+        name: upload macos packages
+        uses: actions/upload-artifact@v3
+        with:
+          name: macos-packages
+          path: dist/netbird_darwin**
+          retention-days: 3
 
   release_ui:
     runs-on: ubuntu-latest
@@ -190,6 +201,9 @@ jobs:
       -
         name: Install modules
         run: go mod tidy
+      - 
+        name: check git status
+        run: git --no-pager diff --exit-code
       -
         name: Run GoReleaser
         id: goreleaser

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

@@ -127,6 +127,9 @@ jobs:
       - name: Install modules
         run: go mod tidy
 
+      - name: check git status
+        run: git --no-pager diff --exit-code
+
       - name: Build management binary
         working-directory: management
         run: CGO_ENABLED=1 go build -o netbird-mgmt main.go
@@ -159,6 +162,13 @@ jobs:
           test $count -eq 4
         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
+
   test-getting-started-script:
     runs-on: ubuntu-latest
     steps:
@@ -186,3 +196,16 @@ jobs:
         run: test -f zitadel.env
       - name: test dashboard.env file gen
         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

.gitignore

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

.golangci.yaml

@@ -63,6 +63,14 @@ linters-settings:
     enable:
       - nilness
 
+  revive:
+    rules:
+      - name: exported
+        severity: warning
+        disabled: false
+        arguments:
+          - "checkPrivateReceivers"
+          - "sayRepetitiveInsteadOfStutters"
   tenv:
     # The option `all` will run against whole test files (`_test.go`) regardless of method/function signatures.
     # Otherwise, only methods that take `*testing.T`, `*testing.B`, and `testing.TB` as arguments are checked.
@@ -93,6 +101,7 @@ linters:
     - nilerr # finds the code that returns nil even if it checks that the error is not nil
     - nilnil # checks that there is no simultaneous return of nil error and an invalid value
     - predeclared # predeclared finds code that shadows one of Go's predeclared identifiers
+    - revive # Fast, configurable, extensible, flexible, and beautiful linter for Go. Drop-in replacement of golint.
     - sqlclosecheck # checks that sql.Rows and sql.Stmt are closed
     - thelper # thelper detects Go test helpers without t.Helper() call and checks the consistency of test helpers.
     - wastedassign # wastedassign finds wasted assignment statements
@@ -121,3 +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_ui.yaml

@@ -54,7 +54,7 @@ nfpms:
     contents:
       - src: client/ui/netbird.desktop
         dst: /usr/share/applications/netbird.desktop
-      - src: client/ui/netbird-systemtray-default.png
+      - src: client/ui/netbird-systemtray-connected.png
         dst: /usr/share/pixmaps/netbird.png
     dependencies:
       - netbird
@@ -71,7 +71,7 @@ nfpms:
     contents:
       - src: client/ui/netbird.desktop
         dst: /usr/share/applications/netbird.desktop
-      - src: client/ui/netbird-systemtray-default.png
+      - src: client/ui/netbird-systemtray-connected.png
         dst: /usr/share/pixmaps/netbird.png
     dependencies:
       - netbird

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.
 

README.md

@@ -1,6 +1,6 @@
 <p align="center">
- <strong>:hatching_chick: New Release! Self-hosting in under 5 min.</strong>
-  <a href="https://github.com/netbirdio/netbird#quickstart-with-self-hosted-netbird">
+ <strong>:hatching_chick: New Release! Device Posture Checks.</strong>
+  <a href="https://docs.netbird.io/how-to/manage-posture-checks">
        Learn more
      </a>   
 </p>
@@ -40,27 +40,26 @@
 
 **Connect.** NetBird creates a WireGuard-based overlay network that automatically connects your machines over an encrypted tunnel, leaving behind the hassle of opening ports, complex firewall rules, VPN gateways, and so forth.
 
-**Secure.** NetBird enables secure remote access by applying granular access policies, while allowing you to manage them intuitively from a single place. Works universally on any infrastructure.
+**Secure.** NetBird enables secure remote access by applying granular access policies while allowing you to manage them intuitively from a single place. Works universally on any infrastructure.
 
-### Secure peer-to-peer VPN with SSO and MFA in minutes
+### Open-Source Network Security in a Single Platform
+
+
+![netbird_2](https://github.com/netbirdio/netbird/assets/700848/46bc3b73-508d-4a0e-bb9a-f465d68646ab)
 
-https://user-images.githubusercontent.com/700848/197345890-2e2cded5-7b7a-436f-a444-94e80dd24f46.mov
 
 ### Key features
 
-| Connectivity                             	                                                                                      | Management                                 	                             | Automation                                    	                            | Platforms    	                        |
-|---------------------------------------------------------------------------------------------------------------------------------|--------------------------------------------------------------------------|----------------------------------------------------------------------------|---------------------------------------|
-| <ul><li> - \[x] Kernel WireGuard </ul></li>                   	                                                                 | <ul><li> - \[x] [Admin Web UI](https://github.com/netbirdio/dashboard) </ul></li>                           	      | <ul><li> - \[x] [Public API](https://docs.netbird.io/api) </ul></li>                                	     | <ul><li> - \[x] Linux </ul></li>    	 |
-| <ul><li> - \[x] Peer-to-peer connections </ul></li>             	                                                               | <ul><li> - \[x] Auto peer discovery and configuration </ul></li>  	      | <ul><li> - \[x] [Setup keys for bulk network provisioning](https://docs.netbird.io/how-to/register-machines-using-setup-keys) </ul></li>  	     | <ul><li> - \[x] Mac </ul></li>      	 |
-| <ul><li> - \[x] Peer-to-peer encryption </ul></li>              	                                                               | <ul><li> - \[x] [IdP integrations](https://docs.netbird.io/selfhosted/identity-providers) </ul></li>                       	      | <ul><li> - \[x] [Self-hosting quickstart script](https://docs.netbird.io/selfhosted/selfhosted-quickstart) </ul></li>              	 | <ul><li> - \[x] Windows </ul></li>  	 |
-| <ul><li> - \[x] Connection relay fallback </ul></li>            	                                                               | <ul><li> - \[x] [SSO & MFA support](https://docs.netbird.io/how-to/installation#running-net-bird-with-sso-login) </ul></li>                      	      | <ul><li> - \[x] IdP groups sync with JWT </ul></li> 	                      | <ul><li> - \[x] Android </ul></li>  	 |
-| <ul><li> - \[x] [Routes to external networks](https://docs.netbird.io/how-to/routing-traffic-to-private-networks) </ul></li>  	 | <ul><li> - \[x] [Access control - groups & rules](https://docs.netbird.io/how-to/manage-network-access) </ul></li>      	        | 	                                                                          | <ul><li> - \[x] iOS </ul></li>      	 |
-| <ul><li> - \[x] NAT traversal with BPF </ul></li>                                                                               | <ul><li> - \[x] [Private DNS](https://docs.netbird.io/how-to/manage-dns-in-your-network) </ul></li>                            	      | 	                                                                          | <ul><li> - \[x] Docker </ul></li>   	 |
-| <ul><li> - \[x] Post-quantum-secure connection through [Rosenpass](https://rosenpass.eu) </ul></li>                             | <ul><li> - \[x] [Multiuser support](https://docs.netbird.io/how-to/add-users-to-your-network) </ul></li>                      	      | 	                                                                          | <ul><li> - \[x] OpenWRT </ul></li>  	 |
-| 	                                                                                                                               | <ul><li> - \[x] [Activity logging](https://docs.netbird.io/how-to/monitor-system-and-network-activity) </ul></li>                       	      | 	                                                                          | 	                                     |
-| 	                                                                                                                               | <ul><li> - \[x] SSH access management </ul></li>                       	 | 	                                                                          | 	                                     |
-
-
+| Connectivity                                                                                                                 | Management                                                                                               | Security                                                                                                                              | Automation                                                                                                                               | Platforms                                                                               |
+|------------------------------------------------------------------------------------------------------------------------------|----------------------------------------------------------------------------------------------------------|---------------------------------------------------------------------------------------------------------------------------------------|------------------------------------------------------------------------------------------------------------------------------------------|-----------------------------------------------------------------------------------------|
+| <ul><li> - \[x] Kernel WireGuard </ul></li>                                                                                  | <ul><li> - \[x] [Admin Web UI](https://github.com/netbirdio/dashboard) </ul></li>                        | <ul><li> - \[x] [SSO & MFA support](https://docs.netbird.io/how-to/installation#running-net-bird-with-sso-login) </ul></li>           | <ul><li> - \[x] [Public API](https://docs.netbird.io/api) </ul></li>                                                                     | <ul><li> - \[x] Linux </ul></li>                                                        |
+| <ul><li> - \[x] Peer-to-peer connections </ul></li>                                                                          | <ul><li> - \[x] Auto peer discovery and configuration </ul></li>                                         | <ul><li> - \[x] [Access control - groups & rules](https://docs.netbird.io/how-to/manage-network-access) </ul></li>                    | <ul><li> - \[x] [Setup keys for bulk network provisioning](https://docs.netbird.io/how-to/register-machines-using-setup-keys) </ul></li> | <ul><li> - \[x] Mac </ul></li>                                                          |
+| <ul><li> - \[x] Connection relay fallback </ul></li>                                                                         | <ul><li> - \[x] [IdP integrations](https://docs.netbird.io/selfhosted/identity-providers) </ul></li>     | <ul><li> - \[x] [Activity logging](https://docs.netbird.io/how-to/monitor-system-and-network-activity) </ul></li>                     | <ul><li> - \[x] [Self-hosting quickstart script](https://docs.netbird.io/selfhosted/selfhosted-quickstart) </ul></li>                    | <ul><li> - \[x] Windows </ul></li>                                                      |
+| <ul><li> - \[x] [Routes to external networks](https://docs.netbird.io/how-to/routing-traffic-to-private-networks) </ul></li> | <ul><li> - \[x] [Private DNS](https://docs.netbird.io/how-to/manage-dns-in-your-network) </ul></li>      | <ul><li> - \[x] [Device posture checks](https://docs.netbird.io/how-to/manage-posture-checks) </ul></li>                              | <ul><li> - \[x] IdP groups sync with JWT </ul></li>                                                                                      | <ul><li> - \[x] Android </ul></li>                                                      |
+| <ul><li> - \[x] NAT traversal with BPF </ul></li>                                                                            | <ul><li> - \[x] [Multiuser support](https://docs.netbird.io/how-to/add-users-to-your-network) </ul></li> | <ul><li> -  \[x] Peer-to-peer encryption </ul></li>                                                                                   |                                                                                                                                          | <ul><li> - \[x] iOS </ul></li>                                                          |
+|                                                                                                                              |                                                                                                          | <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)
@@ -79,7 +78,7 @@ Follow the [Advanced guide with a custom identity provider](https://docs.netbird
 - **Public domain** name pointing to the VM.
 
 **Software requirements:**
-- Docker installed on the VM with the docker compose plugin ([Docker installation guide](https://docs.docker.com/engine/install/)) or docker with docker-compose in version 2 or higher.
+- Docker installed on the VM with the docker-compose plugin ([Docker installation guide](https://docs.docker.com/engine/install/)) or docker with docker-compose in version 2 or higher.
 - [jq](https://jqlang.github.io/jq/) installed. In most distributions
   Usually available in the official repositories and can be installed with `sudo apt install jq` or `sudo yum install jq`
 - [curl](https://curl.se/) installed.
@@ -96,9 +95,9 @@ export NETBIRD_DOMAIN=netbird.example.com; curl -fsSL https://github.com/netbird
 -  Every machine in the network runs [NetBird Agent (or Client)](client/) that manages WireGuard.
 -  Every agent connects to [Management Service](management/) that holds network state, manages peer IPs, and distributes network updates to agents (peers).
 -  NetBird agent uses WebRTC ICE implemented in [pion/ice library](https://github.com/pion/ice) to discover connection candidates when establishing a peer-to-peer connection between machines.
--  Connection candidates are discovered with a help of [STUN](https://en.wikipedia.org/wiki/STUN) servers.
+-  Connection candidates are discovered with the help of [STUN](https://en.wikipedia.org/wiki/STUN) servers.
 -  Agents negotiate a connection through [Signal Service](signal/) passing p2p encrypted messages with candidates.
--  Sometimes the NAT traversal is unsuccessful due to strict NATs (e.g. mobile carrier-grade NAT) and p2p connection isn't possible. When this occurs the system falls back to a relay server called [TURN](https://en.wikipedia.org/wiki/Traversal_Using_Relays_around_NAT), and a secure WireGuard tunnel is established via the TURN server. 
+-  Sometimes the NAT traversal is unsuccessful due to strict NATs (e.g. mobile carrier-grade NAT) and a p2p connection isn't possible. When this occurs the system falls back to a relay server called [TURN](https://en.wikipedia.org/wiki/Traversal_Using_Relays_around_NAT), and a secure WireGuard tunnel is established via the TURN server. 
  
 [Coturn](https://github.com/coturn/coturn) is the one that has been successfully used for STUN and TURN in NetBird setups.
 
@@ -109,8 +108,8 @@ export NETBIRD_DOMAIN=netbird.example.com; curl -fsSL https://github.com/netbird
 See a complete [architecture overview](https://docs.netbird.io/about-netbird/how-netbird-works#architecture) for details.
 
 ### Community projects
--  [NetBird on OpenWRT](https://github.com/messense/openwrt-netbird)
 -  [NetBird installer script](https://github.com/physk/netbird-installer)
+-  [NetBird ansible collection by Dominion Solutions](https://galaxy.ansible.com/ui/repo/published/dominion_solutions/netbird/)
 
 **Note**: The `main` branch may be in an *unstable or even broken state* during development.
 For stable versions, see [releases](https://github.com/netbirdio/netbird/releases).
@@ -122,7 +121,7 @@ In November 2022, NetBird joined the [StartUpSecure program](https://www.forschu
 ![CISPA_Logo_BLACK_EN_RZ_RGB (1)](https://user-images.githubusercontent.com/700848/203091324-c6d311a0-22b5-4b05-a288-91cbc6cdcc46.png)
 
 ### Testimonials
-We use open-source technologies like [WireGuard®](https://www.wireguard.com/), [Pion ICE (WebRTC)](https://github.com/pion/ice), [Coturn](https://github.com/coturn/coturn), and [Rosenpass](https://rosenpass.eu). We very much appreciate the work these guys are doing and we'd greatly appreciate if you could support them in any way (e.g. giving a star or a contribution).
+We use open-source technologies like [WireGuard®](https://www.wireguard.com/), [Pion ICE (WebRTC)](https://github.com/pion/ice), [Coturn](https://github.com/coturn/coturn), and [Rosenpass](https://rosenpass.eu). We very much appreciate the work these guys are doing and we'd greatly appreciate if you could support them in any way (e.g., by giving a star or a contribution).
 
 ### Legal
  _WireGuard_ and the _WireGuard_ logo are [registered trademarks](https://www.wireguard.com/trademark-policy/) of Jason A. Donenfeld.

client/Dockerfile

@@ -1,5 +1,5 @@
 FROM alpine:3.18.5
 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/android/client.go

@@ -1,3 +1,5 @@
+//go:build android
+
 package android
 
 import (
@@ -14,6 +16,7 @@ import (
 	"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
@@ -54,14 +57,17 @@ type Client struct {
 	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(""),
@@ -79,10 +85,14 @@ func (c *Client) Run(urlOpener URLOpener, dns *DNSList, dnsReadyListener DnsRead
 		return err
 	}
 	c.recorder.UpdateManagementAddress(cfg.ManagementURL.String())
+	c.recorder.UpdateRosenpass(cfg.RosenpassEnabled, cfg.RosenpassPermissive)
 
 	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()
@@ -96,7 +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).
@@ -109,6 +120,7 @@ func (c *Client) RunWithoutLogin(dns *DNSList, dnsReadyListener DnsReadyListener
 		return err
 	}
 	c.recorder.UpdateManagementAddress(cfg.ManagementURL.String())
+	c.recorder.UpdateRosenpass(cfg.RosenpassEnabled, cfg.RosenpassPermissive)
 
 	var ctx context.Context
 	//nolint
@@ -120,7 +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
@@ -139,6 +152,11 @@ func (c *Client) SetTraceLogLevel() {
 	log.SetLevel(log.TraceLevel)
 }
 
+// SetInfoLogLevel configure the logger to info level
+func (c *Client) SetInfoLogLevel() {
+	log.SetLevel(log.InfoLevel)
+}
+
 // PeersList return with the list of the PeerInfos
 func (c *Client) PeersList() *PeerInfoArray {
 

client/anonymize/anonymize.go

@@ -0,0 +1,212 @@
+package anonymize
+
+import (
+	"crypto/rand"
+	"fmt"
+	"math/big"
+	"net"
+	"net/netip"
+	"net/url"
+	"regexp"
+	"slices"
+	"strings"
+)
+
+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
+}
+
+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,
+	}
+}
+
+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 {
+	if strings.HasSuffix(domain, "netbird.io") ||
+		strings.HasSuffix(domain, "netbird.selfhosted") ||
+		strings.HasSuffix(domain, "netbird.cloud") ||
+		strings.HasSuffix(domain, "netbird.stage") ||
+		strings.HasSuffix(domain, ".domain") {
+		return domain
+	}
+
+	parts := strings.Split(domain, ".")
+	if len(parts) < 2 {
+		return domain
+	}
+
+	baseDomain := parts[len(parts)-2] + "." + parts[len(parts)-1]
+
+	anonymized, ok := a.domainAnonymizer[baseDomain]
+	if !ok {
+		anonymizedBase := "anon-" + generateRandomString(5) + ".domain"
+		a.domainAnonymizer[baseDomain] = anonymizedBase
+		anonymized = anonymizedBase
+	}
+
+	return strings.Replace(domain, baseDomain, anonymized, 1)
+}
+
+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 stun, stuns, turn, and turns schemes.
+func (a *Anonymizer) AnonymizeSchemeURI(text string) string {
+	re := regexp.MustCompile(`(?i)\b(stuns?:|turns?:|https?://)\S+\b`)
+
+	return re.ReplaceAllStringFunc(text, a.AnonymizeURI)
+}
+
+// AnonymizeDNSLogLine anonymizes domain names in DNS log entries by replacing them with a random string.
+func (a *Anonymizer) AnonymizeDNSLogLine(logEntry string) string {
+	domainPattern := `dns\.Question{Name:"([^"]+)",`
+	domainRegex := regexp.MustCompile(domainPattern)
+
+	return domainRegex.ReplaceAllStringFunc(logEntry, func(match string) string {
+		parts := strings.Split(match, `"`)
+		if len(parts) >= 2 {
+			domain := parts[1]
+			if strings.HasSuffix(domain, ".domain") {
+				return match
+			}
+			randomDomain := generateRandomString(10) + ".domain"
+			return strings.Replace(match, domain, randomDomain, 1)
+		}
+		return match
+	})
+}
+
+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
+	}
+
+	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 +1,223 @@
+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{})
+	testLog := `2024-04-23T20:01:11+02:00 TRAC client/internal/dns/local.go:25: received question: dns.Question{Name:"example.com", Qtype:0x1c, Qclass:0x1}`
+
+	result := anonymizer.AnonymizeDNSLogLine(testLog)
+	require.NotEqual(t, testLog, result)
+	assert.NotContains(t, result, "example.com")
+}
+
+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$`,
+			true,
+		},
+		{
+			"Subdomain",
+			"sub.example.com",
+			`^sub\.anon-[a-zA-Z0-9]+\.domain$`,
+			true,
+		},
+		{
+			"Protected Domain",
+			"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`},
+		{"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`},
+		{"HTTPS URI in message", "Visit https://example.com for more", `Visit https://anon-[a-zA-Z0-9]+\.domain for more`},
+	}
+
+	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 +1,255 @@
+package cmd
+
+import (
+	"context"
+	"fmt"
+	"time"
+
+	"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"
+)
+
+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,
+}
+
+func debugBundle(cmd *cobra.Command, _ []string) error {
+	conn, err := getClient(cmd)
+	if err != nil {
+		return err
+	}
+	defer conn.Close()
+
+	client := proto.NewDaemonServiceClient(conn)
+	resp, err := client.DebugBundle(cmd.Context(), &proto.DebugBundleRequest{
+		Anonymize: anonymizeFlag,
+		Status:    getStatusOutput(cmd),
+	})
+	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 conn.Close()
+
+	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 conn.Close()
+
+	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())
+	}
+
+	restoreUp := 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 _, err := client.Down(cmd.Context(), &proto.DownRequest{}); err != nil {
+		return fmt.Errorf("failed to down: %v", status.Convert(err).Message())
+	}
+	cmd.Println("Netbird down")
+
+	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.")
+	}
+
+	time.Sleep(1 * time.Second)
+
+	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")
+
+	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))
+
+	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)
+
+	if restoreUp {
+		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")
+	}
+
+	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("Creating debug bundle...")
+
+	resp, err := client.DebugBundle(cmd.Context(), &proto.DebugBundleRequest{
+		Anonymize: anonymizeFlag,
+		Status:    statusOutput,
+	})
+	if err != nil {
+		return fmt.Errorf("failed to bundle debug: %v", status.Convert(err).Message())
+	}
+
+	cmd.Println(resp.GetPath())
+
+	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"
 

client/cmd/root.go

@@ -25,12 +25,17 @@ import (
 )
 
 const (
-	externalIPMapFlag   = "external-ip-map"
-	dnsResolverAddress  = "dns-resolver-address"
-	enableRosenpassFlag = "enable-rosenpass"
-	preSharedKeyFlag    = "preshared-key"
-	interfaceNameFlag   = "interface-name"
-	wireguardPortFlag   = "wireguard-port"
+	externalIPMapFlag       = "external-ip-map"
+	dnsResolverAddress      = "dns-resolver-address"
+	enableRosenpassFlag     = "enable-rosenpass"
+	rosenpassPermissiveFlag = "rosenpass-permissive"
+	preSharedKeyFlag        = "preshared-key"
+	interfaceNameFlag       = "interface-name"
+	wireguardPortFlag       = "wireguard-port"
+	networkMonitorFlag      = "network-monitor"
+	disableAutoConnectFlag  = "disable-auto-connect"
+	serverSSHAllowedFlag    = "allow-server-ssh"
+	extraIFaceBlackListFlag = "extra-iface-blacklist"
 )
 
 var (
@@ -54,8 +59,15 @@ var (
 	natExternalIPs          []string
 	customDNSAddress        string
 	rosenpassEnabled        bool
+	rosenpassPermissive     bool
+	serverSSHAllowed        bool
 	interfaceName           string
 	wireguardPort           uint16
+	networkMonitor          bool
+	serviceName             string
+	autoConnectDisabled     bool
+	extraIFaceBlackList     []string
+	anonymizeFlag           bool
 	rootCmd                 = &cobra.Command{
 		Use:          "netbird",
 		Short:        "",
@@ -94,15 +106,24 @@ func init() {
 	if runtime.GOOS == "windows" {
 		defaultDaemonAddr = "tcp://127.0.0.1:41731"
 	}
+
+	defaultServiceName := "netbird"
+	if runtime.GOOS == "windows" {
+		defaultServiceName = "Netbird"
+	}
+
 	rootCmd.PersistentFlags().StringVar(&daemonAddr, "daemon-addr", defaultDaemonAddr, "Daemon service address to serve CLI requests [unix|tcp]://[path|host:port]")
 	rootCmd.PersistentFlags().StringVarP(&managementURL, "management-url", "m", "", fmt.Sprintf("Management Service URL [http|https]://[host]:[port] (default \"%s\")", internal.DefaultManagementURL))
 	rootCmd.PersistentFlags().StringVar(&adminURL, "admin-url", "", fmt.Sprintf("Admin Panel URL [http|https]://[host]:[port] (default \"%s\")", internal.DefaultAdminURL))
+	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().StringVarP(&setupKey, "setup-key", "k", "", "Setup key obtained from the Management Service Dashboard (used to register peer)")
 	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)
@@ -110,8 +131,20 @@ func init() {
 	rootCmd.AddCommand(loginCmd)
 	rootCmd.AddCommand(versionCmd)
 	rootCmd.AddCommand(sshCmd)
+	rootCmd.AddCommand(routesCmd)
+	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
+
+	routesCmd.AddCommand(routesListCmd)
+	routesCmd.AddCommand(routesSelectCmd, routesDeselectCmd)
+
+	debugCmd.AddCommand(debugBundleCmd)
+	debugCmd.AddCommand(logCmd)
+	logCmd.AddCommand(logLevelCmd)
+	debugCmd.AddCommand(forCmd)
+
 	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. `+
@@ -126,6 +159,9 @@ func init() {
 			`E.g. --dns-resolver-address 127.0.0.1:5053 or --dns-resolver-address ""`,
 	)
 	upCmd.PersistentFlags().BoolVar(&rosenpassEnabled, enableRosenpassFlag, false, "[Experimental] Enable Rosenpass feature. If enabled, the connection will be post-quantum secured via Rosenpass.")
+	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.")
 }
 
 // SetupCloseHandler handles SIGTERM signal and exits with success
@@ -176,7 +212,7 @@ func FlagNameToEnvVar(cmdFlag string, prefix string) string {
 	return prefix + upper
 }
 
-// DialClientGRPCServer returns client connection to the dameno server.
+// DialClientGRPCServer returns client connection to the daemon server.
 func DialClientGRPCServer(ctx context.Context, addr string) (*grpc.ClientConn, error) {
 	ctx, cancel := context.WithTimeout(ctx, time.Second*3)
 	defer cancel()
@@ -316,3 +352,17 @@ 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/route.go

@@ -0,0 +1,131 @@
+package cmd
+
+import (
+	"fmt"
+
+	"github.com/spf13/cobra"
+	"google.golang.org/grpc/status"
+
+	"github.com/netbirdio/netbird/client/proto"
+)
+
+var appendFlag bool
+
+var routesCmd = &cobra.Command{
+	Use:   "routes",
+	Short: "Manage network routes",
+	Long:  `Commands to list, select, or deselect network routes.`,
+}
+
+var routesListCmd = &cobra.Command{
+	Use:     "list",
+	Aliases: []string{"ls"},
+	Short:   "List routes",
+	Example: "  netbird routes list",
+	Long:    "List all available network routes.",
+	RunE:    routesList,
+}
+
+var routesSelectCmd = &cobra.Command{
+	Use:     "select route...|all",
+	Short:   "Select routes",
+	Long:    "Select a list of routes by identifiers or 'all' to clear all selections and to accept all (including new) routes.\nDefault mode is replace, use -a to append to already selected routes.",
+	Example: "  netbird routes select all\n  netbird routes select route1 route2\n  netbird routes select -a route3",
+	Args:    cobra.MinimumNArgs(1),
+	RunE:    routesSelect,
+}
+
+var routesDeselectCmd = &cobra.Command{
+	Use:     "deselect route...|all",
+	Short:   "Deselect routes",
+	Long:    "Deselect previously selected routes by identifiers or 'all' to disable accepting any routes.",
+	Example: "  netbird routes deselect all\n  netbird routes deselect route1 route2",
+	Args:    cobra.MinimumNArgs(1),
+	RunE:    routesDeselect,
+}
+
+func init() {
+	routesSelectCmd.PersistentFlags().BoolVarP(&appendFlag, "append", "a", false, "Append to current route selection instead of replacing")
+}
+
+func routesList(cmd *cobra.Command, _ []string) error {
+	conn, err := getClient(cmd)
+	if err != nil {
+		return err
+	}
+	defer conn.Close()
+
+	client := proto.NewDaemonServiceClient(conn)
+	resp, err := client.ListRoutes(cmd.Context(), &proto.ListRoutesRequest{})
+	if err != nil {
+		return fmt.Errorf("failed to list routes: %v", status.Convert(err).Message())
+	}
+
+	if len(resp.Routes) == 0 {
+		cmd.Println("No routes available.")
+		return nil
+	}
+
+	cmd.Println("Available Routes:")
+	for _, route := range resp.Routes {
+		selectedStatus := "Not Selected"
+		if route.GetSelected() {
+			selectedStatus = "Selected"
+		}
+		cmd.Printf("\n  - ID: %s\n    Network: %s\n    Status: %s\n", route.GetID(), route.GetNetwork(), selectedStatus)
+	}
+
+	return nil
+}
+
+func routesSelect(cmd *cobra.Command, args []string) error {
+	conn, err := getClient(cmd)
+	if err != nil {
+		return err
+	}
+	defer conn.Close()
+
+	client := proto.NewDaemonServiceClient(conn)
+	req := &proto.SelectRoutesRequest{
+		RouteIDs: args,
+	}
+
+	if len(args) == 1 && args[0] == "all" {
+		req.All = true
+	} else if appendFlag {
+		req.Append = true
+	}
+
+	if _, err := client.SelectRoutes(cmd.Context(), req); err != nil {
+		return fmt.Errorf("failed to select routes: %v", status.Convert(err).Message())
+	}
+
+	cmd.Println("Routes selected successfully.")
+
+	return nil
+}
+
+func routesDeselect(cmd *cobra.Command, args []string) error {
+	conn, err := getClient(cmd)
+	if err != nil {
+		return err
+	}
+	defer conn.Close()
+
+	client := proto.NewDaemonServiceClient(conn)
+	req := &proto.SelectRoutesRequest{
+		RouteIDs: args,
+	}
+
+	if len(args) == 1 && args[0] == "all" {
+		req.All = true
+	}
+
+	if _, err := client.DeselectRoutes(cmd.Context(), req); err != nil {
+		return fmt.Errorf("failed to deselect routes: %v", status.Convert(err).Message())
+	}
+
+	cmd.Println("Routes deselected successfully.")
+
+	return nil
+}

client/cmd/service.go

@@ -2,8 +2,6 @@ package cmd
 
 import (
 	"context"
-	"runtime"
-
 	"github.com/kardianos/service"
 	log "github.com/sirupsen/logrus"
 	"github.com/spf13/cobra"
@@ -24,12 +22,8 @@ func newProgram(ctx context.Context, cancel context.CancelFunc) *program {
 }
 
 func newSVCConfig() *service.Config {
-	name := "netbird"
-	if runtime.GOOS == "windows" {
-		name = "Netbird"
-	}
 	return &service.Config{
-		Name:        name,
+		Name:        serviceName,
 		DisplayName: "Netbird",
 		Description: "A WireGuard-based mesh network that connects your devices into a single private network.",
 		Option:      make(service.KeyValue),

client/cmd/service_controller.go

@@ -11,11 +11,12 @@ import (
 	"github.com/kardianos/service"
 	log "github.com/sirupsen/logrus"
 
+	"github.com/spf13/cobra"
+	"google.golang.org/grpc"
+
 	"github.com/netbirdio/netbird/client/proto"
 	"github.com/netbirdio/netbird/client/server"
 	"github.com/netbirdio/netbird/util"
-	"github.com/spf13/cobra"
-	"google.golang.org/grpc"
 )
 
 func (p *program) Start(svc service.Service) error {
@@ -109,7 +110,6 @@ var runCmd = &cobra.Command{
 		if err != nil {
 			return err
 		}
-		cmd.Printf("Netbird service is running")
 		return nil
 	},
 }

client/cmd/service_installer.go

@@ -64,6 +64,10 @@ var installCmd = &cobra.Command{
 			}
 		}
 
+		if runtime.GOOS == "windows" {
+			svcConfig.Option["OnFailure"] = "restart"
+		}
+
 		ctx, cancel := context.WithCancel(cmd.Context())
 
 		s, err := newSVC(newProgram(ctx, cancel), svcConfig)
@@ -77,6 +81,7 @@ var installCmd = &cobra.Command{
 			cmd.PrintErrln(err)
 			return err
 		}
+
 		cmd.Println("Netbird service has been installed")
 		return nil
 	},
@@ -106,7 +111,7 @@ var uninstallCmd = &cobra.Command{
 		if err != nil {
 			return err
 		}
-		cmd.Println("Netbird has been uninstalled")
+		cmd.Println("Netbird service has been uninstalled")
 		return nil
 	},
 }

client/cmd/ssh.go

@@ -24,7 +24,7 @@ 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")
@@ -94,7 +94,7 @@ 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/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"
@@ -34,6 +37,9 @@ type peerStateDetailOutput struct {
 	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"`
 }
 
 type peersStateOutput struct {
@@ -71,17 +77,28 @@ type iceCandidateType struct {
 	Remote string `json:"remote" yaml:"remote"`
 }
 
+type nsServerGroupStateOutput struct {
+	Servers []string `json:"servers" yaml:"servers"`
+	Domains []string `json:"domains" yaml:"domains"`
+	Enabled bool     `json:"enabled" yaml:"enabled"`
+	Error   string   `json:"error" yaml:"error"`
+}
+
 type statusOutputOverview struct {
-	Peers           peersStateOutput      `json:"peers" yaml:"peers"`
-	CliVersion      string                `json:"cliVersion" yaml:"cliVersion"`
-	DaemonVersion   string                `json:"daemonVersion" yaml:"daemonVersion"`
-	ManagementState managementStateOutput `json:"management" yaml:"management"`
-	SignalState     signalStateOutput     `json:"signal" yaml:"signal"`
-	Relays          relayStateOutput      `json:"relays" yaml:"relays"`
-	IP              string                `json:"netbirdIp" yaml:"netbirdIp"`
-	PubKey          string                `json:"publicKey" yaml:"publicKey"`
-	KernelInterface bool                  `json:"usesKernelInterface" yaml:"usesKernelInterface"`
-	FQDN            string                `json:"fqdn" yaml:"fqdn"`
+	Peers               peersStateOutput           `json:"peers" yaml:"peers"`
+	CliVersion          string                     `json:"cliVersion" yaml:"cliVersion"`
+	DaemonVersion       string                     `json:"daemonVersion" yaml:"daemonVersion"`
+	ManagementState     managementStateOutput      `json:"management" yaml:"management"`
+	SignalState         signalStateOutput          `json:"signal" yaml:"signal"`
+	Relays              relayStateOutput           `json:"relays" yaml:"relays"`
+	IP                  string                     `json:"netbirdIp" yaml:"netbirdIp"`
+	PubKey              string                     `json:"publicKey" yaml:"publicKey"`
+	KernelInterface     bool                       `json:"usesKernelInterface" yaml:"usesKernelInterface"`
+	FQDN                string                     `json:"fqdn" yaml:"fqdn"`
+	RosenpassEnabled    bool                       `json:"quantumResistance" yaml:"quantumResistance"`
+	RosenpassPermissive bool                       `json:"quantumResistancePermissive" yaml:"quantumResistancePermissive"`
+	Routes              []string                   `json:"routes" yaml:"routes"`
+	NSServerGroups      []nsServerGroupStateOutput `json:"dnsServers" yaml:"dnsServers"`
 }
 
 var (
@@ -130,9 +147,9 @@ 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
 	}
@@ -165,7 +182,7 @@ func statusFunc(cmd *cobra.Command, args []string) error {
 	case yamlFlag:
 		statusOutputString, err = parseToYAML(outputInformationHolder)
 	default:
-		statusOutputString = parseGeneralSummary(outputInformationHolder, false, false)
+		statusOutputString = parseGeneralSummary(outputInformationHolder, false, false, false)
 	}
 
 	if err != nil {
@@ -177,7 +194,7 @@ 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"+
@@ -186,7 +203,7 @@ 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())
 	}
@@ -253,16 +270,25 @@ func convertToStatusOutputOverview(resp *proto.StatusResponse) statusOutputOverv
 	peersOverview := mapPeers(resp.GetFullStatus().GetPeers())
 
 	overview := statusOutputOverview{
-		Peers:           peersOverview,
-		CliVersion:      version.NetbirdVersion(),
-		DaemonVersion:   resp.GetDaemonVersion(),
-		ManagementState: managementOverview,
-		SignalState:     signalOverview,
-		Relays:          relayOverview,
-		IP:              pbFullStatus.GetLocalPeerState().GetIP(),
-		PubKey:          pbFullStatus.GetLocalPeerState().GetPubKey(),
-		KernelInterface: pbFullStatus.GetLocalPeerState().GetKernelInterface(),
-		FQDN:            pbFullStatus.GetLocalPeerState().GetFqdn(),
+		Peers:               peersOverview,
+		CliVersion:          version.NetbirdVersion(),
+		DaemonVersion:       resp.GetDaemonVersion(),
+		ManagementState:     managementOverview,
+		SignalState:         signalOverview,
+		Relays:              relayOverview,
+		IP:                  pbFullStatus.GetLocalPeerState().GetIP(),
+		PubKey:              pbFullStatus.GetLocalPeerState().GetPubKey(),
+		KernelInterface:     pbFullStatus.GetLocalPeerState().GetKernelInterface(),
+		FQDN:                pbFullStatus.GetLocalPeerState().GetFqdn(),
+		RosenpassEnabled:    pbFullStatus.GetLocalPeerState().GetRosenpassEnabled(),
+		RosenpassPermissive: pbFullStatus.GetLocalPeerState().GetRosenpassPermissive(),
+		Routes:              pbFullStatus.GetLocalPeerState().GetRoutes(),
+		NSServerGroups:      mapNSGroups(pbFullStatus.GetDnsServers()),
+	}
+
+	if anonymizeFlag {
+		anonymizer := anonymize.NewAnonymizer(anonymize.DefaultAddresses())
+		anonymizeOverview(anonymizer, &overview)
 	}
 
 	return overview
@@ -294,6 +320,19 @@ func mapRelays(relays []*proto.RelayState) relayStateOutput {
 	}
 }
 
+func mapNSGroups(servers []*proto.NSGroupState) []nsServerGroupStateOutput {
+	mappedNSGroups := make([]nsServerGroupStateOutput, 0, len(servers))
+	for _, pbNsGroupServer := range servers {
+		mappedNSGroups = append(mappedNSGroups, nsServerGroupStateOutput{
+			Servers: pbNsGroupServer.GetServers(),
+			Domains: pbNsGroupServer.GetDomains(),
+			Enabled: pbNsGroupServer.GetEnabled(),
+			Error:   pbNsGroupServer.GetError(),
+		})
+	}
+	return mappedNSGroups
+}
+
 func mapPeers(peers []*proto.PeerState) peersStateOutput {
 	var peersStateDetail []peerStateDetailOutput
 	localICE := ""
@@ -346,6 +385,9 @@ func mapPeers(peers []*proto.PeerState) peersStateOutput {
 			LastWireguardHandshake: lastHandshake,
 			TransferReceived:       transferReceived,
 			TransferSent:           transferSent,
+			Latency:                pbPeerState.GetLatency().AsDuration(),
+			RosenpassEnabled:       pbPeerState.GetRosenpassEnabled(),
+			Routes:                 pbPeerState.GetRoutes(),
 		}
 
 		peersStateDetail = append(peersStateDetail, peerState)
@@ -395,8 +437,7 @@ func parseToYAML(overview statusOutputOverview) (string, error) {
 	return string(yamlBytes), nil
 }
 
-func parseGeneralSummary(overview statusOutputOverview, showURL bool, showRelays bool) string {
-
+func parseGeneralSummary(overview statusOutputOverview, showURL bool, showRelays bool, showNameServers bool) string {
 	var managementConnString string
 	if overview.ManagementState.Connected {
 		managementConnString = "Connected"
@@ -432,7 +473,7 @@ func parseGeneralSummary(overview statusOutputOverview, showURL bool, showRelays
 		interfaceIP = "N/A"
 	}
 
-	var relayAvailableString string
+	var relaysString string
 	if showRelays {
 		for _, relay := range overview.Relays.Details {
 			available := "Available"
@@ -441,42 +482,98 @@ func parseGeneralSummary(overview statusOutputOverview, showURL bool, showRelays
 				available = "Unavailable"
 				reason = fmt.Sprintf(", reason: %s", relay.Error)
 			}
-			relayAvailableString += fmt.Sprintf("\n  [%s] is %s%s", relay.URI, available, reason)
-
+			relaysString += fmt.Sprintf("\n  [%s] is %s%s", relay.URI, available, reason)
 		}
 	} else {
+		relaysString = fmt.Sprintf("%d/%d Available", overview.Relays.Available, overview.Relays.Total)
+	}
 
-		relayAvailableString = 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, ", ")
+	}
+
+	var dnsServersString string
+	if showNameServers {
+		for _, nsServerGroup := range overview.NSServerGroups {
+			enabled := "Available"
+			if !nsServerGroup.Enabled {
+				enabled = "Unavailable"
+			}
+			errorString := ""
+			if nsServerGroup.Error != "" {
+				errorString = fmt.Sprintf(", reason: %s", nsServerGroup.Error)
+				errorString = strings.TrimSpace(errorString)
+			}
+
+			domainsString := strings.Join(nsServerGroup.Domains, ", ")
+			if domainsString == "" {
+				domainsString = "." // Show "." for the default zone
+			}
+			dnsServersString += fmt.Sprintf(
+				"\n  [%s] for [%s] is %s%s",
+				strings.Join(nsServerGroup.Servers, ", "),
+				domainsString,
+				enabled,
+				errorString,
+			)
+		}
+	} else {
+		dnsServersString = fmt.Sprintf("%d/%d Available", countEnabled(overview.NSServerGroups), len(overview.NSServerGroups))
+	}
+
+	rosenpassEnabledStatus := "false"
+	if overview.RosenpassEnabled {
+		rosenpassEnabledStatus = "true"
+		if overview.RosenpassPermissive {
+			rosenpassEnabledStatus = "true (permissive)" //nolint:gosec
+		}
 	}
 
 	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"+
 			"Relays: %s\n"+
+			"Nameservers: %s\n"+
 			"FQDN: %s\n"+
 			"NetBird IP: %s\n"+
 			"Interface type: %s\n"+
+			"Quantum resistance: %s\n"+
+			"Routes: %s\n"+
 			"Peers count: %s\n",
+		fmt.Sprintf("%s/%s%s", goos, goarch, goarm),
 		overview.DaemonVersion,
 		version.NetbirdVersion(),
 		managementConnString,
 		signalConnString,
-		relayAvailableString,
+		relaysString,
+		dnsServersString,
 		overview.FQDN,
 		interfaceIP,
 		interfaceTypeString,
+		rosenpassEnabledStatus,
+		routes,
 		peersCountString,
 	)
 	return summary
 }
 
 func parseToFullDetailSummary(overview statusOutputOverview) string {
-	parsedPeersString := parsePeers(overview.Peers)
-	summary := parseGeneralSummary(overview, true, true)
+	parsedPeersString := parsePeers(overview.Peers, overview.RosenpassEnabled, overview.RosenpassPermissive)
+	summary := parseGeneralSummary(overview, true, true, true)
 
 	return fmt.Sprintf(
 		"Peers detail:"+
@@ -487,7 +584,7 @@ func parseToFullDetailSummary(overview statusOutputOverview) string {
 	)
 }
 
-func parsePeers(peers peersStateOutput) string {
+func parsePeers(peers peersStateOutput, rosenpassEnabled, rosenpassPermissive bool) string {
 	var (
 		peersString = ""
 	)
@@ -513,14 +610,28 @@ func parsePeers(peers peersStateOutput) string {
 		if peerState.IceCandidateEndpoint.Remote != "" {
 			remoteICEEndpoint = peerState.IceCandidateEndpoint.Remote
 		}
-		lastStatusUpdate := "-"
-		if !peerState.LastStatusUpdate.IsZero() {
-			lastStatusUpdate = peerState.LastStatusUpdate.Format("2006-01-02 15:04:05")
+
+		rosenpassEnabledStatus := "false"
+		if rosenpassEnabled {
+			if peerState.RosenpassEnabled {
+				rosenpassEnabledStatus = "true"
+			} else {
+				if rosenpassPermissive {
+					rosenpassEnabledStatus = "false (remote didn't enable quantum resistance)"
+				} else {
+					rosenpassEnabledStatus = "false (connection won't work without a permissive mode)"
+				}
+			}
+		} else {
+			if peerState.RosenpassEnabled {
+				rosenpassEnabledStatus = "false (connection might not work without a remote permissive mode)"
+			}
 		}
 
-		lastWireguardHandshake := "-"
-		if !peerState.LastWireguardHandshake.IsZero() && peerState.LastWireguardHandshake != time.Unix(0, 0) {
-			lastWireguardHandshake = peerState.LastWireguardHandshake.Format("2006-01-02 15:04:05")
+		routes := "-"
+		if len(peerState.Routes) > 0 {
+			sort.Strings(peerState.Routes)
+			routes = strings.Join(peerState.Routes, ", ")
 		}
 
 		peerString := fmt.Sprintf(
@@ -534,8 +645,11 @@ func parsePeers(peers peersStateOutput) string {
 				"  ICE candidate (Local/Remote): %s/%s\n"+
 				"  ICE candidate endpoints (Local/Remote): %s/%s\n"+
 				"  Last connection update: %s\n"+
-				"  Last Wireguard handshake: %s\n"+
-				"  Transfer status (received/sent) %s/%s\n",
+				"  Last WireGuard handshake: %s\n"+
+				"  Transfer status (received/sent) %s/%s\n"+
+				"  Quantum resistance: %s\n"+
+				"  Routes: %s\n"+
+				"  Latency: %s\n",
 			peerState.FQDN,
 			peerState.IP,
 			peerState.PubKey,
@@ -546,10 +660,13 @@ func parsePeers(peers peersStateOutput) string {
 			remoteICE,
 			localICEEndpoint,
 			remoteICEEndpoint,
-			lastStatusUpdate,
-			lastWireguardHandshake,
+			timeAgo(peerState.LastStatusUpdate),
+			timeAgo(peerState.LastWireguardHandshake),
 			toIEC(peerState.TransferReceived),
 			toIEC(peerState.TransferSent),
+			rosenpassEnabledStatus,
+			routes,
+			peerState.Latency.String(),
 		)
 
 		peersString += peerString
@@ -603,3 +720,139 @@ func toIEC(b int64) string {
 	return fmt.Sprintf("%.1f %ciB",
 		float64(b)/float64(div), "KMGTPE"[exp])
 }
+
+func countEnabled(dnsServers []nsServerGroupStateOutput) int {
+	count := 0
+	for _, server := range dnsServers {
+		if server.Enabled {
+			count++
+		}
+	}
+	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)
+	}
+	for i, route := range peer.Routes {
+		peer.Routes[i] = a.AnonymizeIPString(route)
+	}
+
+	for i, route := range peer.Routes {
+		prefix, err := netip.ParsePrefix(route)
+		if err == nil {
+			ip := a.AnonymizeIPString(prefix.Addr().String())
+			peer.Routes[i] = fmt.Sprintf("%s/%d", ip, prefix.Bits())
+		}
+	}
+}
+
+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.Routes {
+		prefix, err := netip.ParsePrefix(route)
+		if err == nil {
+			ip := a.AnonymizeIPString(prefix.Addr().String())
+			overview.Routes[i] = fmt.Sprintf("%s/%d", ip, prefix.Bits())
+		}
+	}
+
+	overview.FQDN = a.AnonymizeDomain(overview.FQDN)
+}

client/cmd/status_test.go

@@ -3,11 +3,14 @@ package cmd
 import (
 	"bytes"
 	"encoding/json"
+	"fmt"
+	"runtime"
 	"testing"
 	"time"
 
 	"github.com/stretchr/testify/assert"
 	"github.com/stretchr/testify/require"
+	"google.golang.org/protobuf/types/known/durationpb"
 	"google.golang.org/protobuf/types/known/timestamppb"
 
 	"github.com/netbirdio/netbird/client/proto"
@@ -42,6 +45,10 @@ 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{
+					"10.1.0.0/24",
+				},
+				Latency: durationpb.New(time.Duration(10000000)),
 			},
 			{
 				IP:                         "192.168.178.102",
@@ -58,6 +65,7 @@ var resp = &proto.StatusResponse{
 				LastWireguardHandshake:     timestamppb.New(time.Date(2002, time.Month(2), 2, 2, 2, 3, 0, time.UTC)),
 				BytesRx:                    2000,
 				BytesTx:                    1000,
+				Latency:                    durationpb.New(time.Duration(10000000)),
 			},
 		},
 		ManagementState: &proto.ManagementState{
@@ -87,6 +95,31 @@ var resp = &proto.StatusResponse{
 			PubKey:          "Some-Pub-Key",
 			KernelInterface: true,
 			Fqdn:            "some-localhost.awesome-domain.com",
+			Routes: []string{
+				"10.10.0.0/24",
+			},
+		},
+		DnsServers: []*proto.NSGroupState{
+			{
+				Servers: []string{
+					"8.8.8.8:53",
+				},
+				Domains: nil,
+				Enabled: true,
+				Error:   "",
+			},
+			{
+				Servers: []string{
+					"1.1.1.1:53",
+					"2.2.2.2:53",
+				},
+				Domains: []string{
+					"example.com",
+					"example.net",
+				},
+				Enabled: false,
+				Error:   "timeout",
+			},
 		},
 	},
 	DaemonVersion: "0.14.1",
@@ -116,6 +149,10 @@ var overview = statusOutputOverview{
 				LastWireguardHandshake: time.Date(2001, 1, 1, 1, 1, 2, 0, time.UTC),
 				TransferReceived:       200,
 				TransferSent:           100,
+				Routes: []string{
+					"10.1.0.0/24",
+				},
+				Latency: time.Duration(10000000),
 			},
 			{
 				IP:               "192.168.178.102",
@@ -136,6 +173,7 @@ var overview = statusOutputOverview{
 				LastWireguardHandshake: time.Date(2002, 2, 2, 2, 2, 3, 0, time.UTC),
 				TransferReceived:       2000,
 				TransferSent:           1000,
+				Latency:                time.Duration(10000000),
 			},
 		},
 	},
@@ -171,6 +209,31 @@ var overview = statusOutputOverview{
 	PubKey:          "Some-Pub-Key",
 	KernelInterface: true,
 	FQDN:            "some-localhost.awesome-domain.com",
+	NSServerGroups: []nsServerGroupStateOutput{
+		{
+			Servers: []string{
+				"8.8.8.8:53",
+			},
+			Domains: nil,
+			Enabled: true,
+			Error:   "",
+		},
+		{
+			Servers: []string{
+				"1.1.1.1:53",
+				"2.2.2.2:53",
+			},
+			Domains: []string{
+				"example.com",
+				"example.net",
+			},
+			Enabled: false,
+			Error:   "timeout",
+		},
+	},
+	Routes: []string{
+		"10.10.0.0/24",
+	},
 }
 
 func TestConversionFromFullStatusToOutputOverview(t *testing.T) {
@@ -231,7 +294,12 @@ func TestParsingToJSON(t *testing.T) {
                 },
                 "lastWireguardHandshake": "2001-01-01T01:01:02Z",
                 "transferReceived": 200,
-                "transferSent": 100
+                "transferSent": 100,
+				"latency": 10000000,
+                "quantumResistance": false,
+                "routes": [
+                  "10.1.0.0/24"
+                ]
               },
               {
                 "fqdn": "peer-2.awesome-domain.com",
@@ -251,7 +319,10 @@ func TestParsingToJSON(t *testing.T) {
                 },
                 "lastWireguardHandshake": "2002-02-02T02:02:03Z",
                 "transferReceived": 2000,
-                "transferSent": 1000
+                "transferSent": 1000,
+				"latency": 10000000,
+                "quantumResistance": false,
+                "routes": null
               }
             ]
           },
@@ -286,7 +357,34 @@ func TestParsingToJSON(t *testing.T) {
           "netbirdIp": "192.168.178.100/16",
           "publicKey": "Some-Pub-Key",
           "usesKernelInterface": true,
-          "fqdn": "some-localhost.awesome-domain.com"
+          "fqdn": "some-localhost.awesome-domain.com",
+          "quantumResistance": false,
+          "quantumResistancePermissive": false,
+          "routes": [
+            "10.10.0.0/24"
+          ],
+          "dnsServers": [
+            {
+              "servers": [
+                "8.8.8.8:53"
+              ],
+              "domains": null,
+              "enabled": true,
+              "error": ""
+            },
+            {
+              "servers": [
+                "1.1.1.1:53",
+                "2.2.2.2:53"
+              ],
+              "domains": [
+                "example.com",
+                "example.net"
+              ],
+              "enabled": false,
+              "error": "timeout"
+            }
+          ]
         }`
 	// @formatter:on
 
@@ -320,6 +418,10 @@ func TestParsingToYAML(t *testing.T) {
           lastWireguardHandshake: 2001-01-01T01:01:02Z
           transferReceived: 200
           transferSent: 100
+          latency: 10ms
+          quantumResistance: false
+          routes:
+            - 10.1.0.0/24
         - fqdn: peer-2.awesome-domain.com
           netbirdIp: 192.168.178.102
           publicKey: Pubkey2
@@ -336,6 +438,9 @@ func TestParsingToYAML(t *testing.T) {
           lastWireguardHandshake: 2002-02-02T02:02:03Z
           transferReceived: 2000
           transferSent: 1000
+          latency: 10ms
+          quantumResistance: false
+          routes: []
 cliVersion: development
 daemonVersion: 0.14.1
 management:
@@ -360,15 +465,39 @@ netbirdIp: 192.168.178.100/16
 publicKey: Some-Pub-Key
 usesKernelInterface: true
 fqdn: some-localhost.awesome-domain.com
+quantumResistance: false
+quantumResistancePermissive: false
+routes:
+    - 10.10.0.0/24
+dnsServers:
+    - servers:
+        - 8.8.8.8:53
+      domains: []
+      enabled: true
+      error: ""
+    - servers:
+        - 1.1.1.1:53
+        - 2.2.2.2:53
+      domains:
+        - example.com
+        - example.net
+      enabled: false
+      error: timeout
 `
 
 	assert.Equal(t, expectedYAML, yaml)
 }
 
 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
@@ -379,9 +508,12 @@ func TestParsingToDetail(t *testing.T) {
   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
+  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
+  Latency: 10ms
 
  peer-2.awesome-domain.com:
   NetBird IP: 192.168.178.102
@@ -392,38 +524,50 @@ func TestParsingToDetail(t *testing.T) {
   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
+  Last connection update: %s
+  Last WireGuard handshake: %s
   Transfer status (received/sent) 2.0 KiB/1000 B
+  Quantum resistance: false
+  Routes: -
+  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: 
   [stun:my-awesome-stun.com:3478] is Available
   [turns:my-awesome-turn.com:443?transport=tcp] is Unavailable, reason: context: deadline exceeded
+Nameservers: 
+  [8.8.8.8:53] for [.] is Available
+  [1.1.1.1:53, 2.2.2.2:53] for [example.com, example.net] is Unavailable, reason: timeout
 FQDN: some-localhost.awesome-domain.com
 NetBird IP: 192.168.178.100/16
 Interface type: Kernel
+Quantum resistance: false
+Routes: 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)
 }
 
 func TestParsingToShortVersion(t *testing.T) {
-	shortVersion := parseGeneralSummary(overview, false, false)
+	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
 Relays: 1/2 Available
+Nameservers: 1/2 Available
 FQDN: some-localhost.awesome-domain.com
 NetBird IP: 192.168.178.100/16
 Interface type: Kernel
+Quantum resistance: false
+Routes: 10.10.0.0/24
 Peers count: 2/2 Connected
 `
 
@@ -437,3 +581,31 @@ 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/testutil.go

@@ -13,6 +13,8 @@ import (
 
 	"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"
@@ -68,18 +70,19 @@ func startManagement(t *testing.T, config *mgmt.Config) (*grpc.Server, net.Liste
 		t.Fatal(err)
 	}
 	s := grpc.NewServer()
-	store, err := mgmt.NewStoreFromJson(config.Datadir, nil)
+	store, cleanUp, err := mgmt.NewTestStoreFromJson(config.Datadir)
 	if err != nil {
 		t.Fatal(err)
 	}
+	t.Cleanup(cleanUp)
 
 	peersUpdateManager := mgmt.NewPeersUpdateManager(nil)
 	eventStore := &activity.InMemoryEventStore{}
 	if err != nil {
 		return nil, nil
 	}
-	accountManager, err := mgmt.BuildManager(store, peersUpdateManager, nil, "", "",
-		eventStore, false)
+	iv, _ := integrations.NewIntegratedValidator(eventStore)
+	accountManager, err := mgmt.BuildManager(store, peersUpdateManager, nil, "", "netbird.selfhosted", eventStore, nil, false, iv)
 	if err != nil {
 		t.Fatal(err)
 	}

client/cmd/up.go

@@ -40,6 +40,8 @@ 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", false, "Enable network monitoring")
+	upCmd.PersistentFlags().StringSliceVar(&extraIFaceBlackList, extraIFaceBlackListFlag, nil, "Extra list of default interfaces to ignore for listening")
 }
 
 func upFunc(cmd *cobra.Command, args []string) error {
@@ -83,17 +85,26 @@ func runInForegroundMode(ctx context.Context, cmd *cobra.Command) error {
 	}
 
 	ic := internal.ConfigInput{
-		ManagementURL:    managementURL,
-		AdminURL:         adminURL,
-		ConfigPath:       configPath,
-		NATExternalIPs:   natExternalIPs,
-		CustomDNSAddress: customDNSAddressConverted,
+		ManagementURL:       managementURL,
+		AdminURL:            adminURL,
+		ConfigPath:          configPath,
+		NATExternalIPs:      natExternalIPs,
+		CustomDNSAddress:    customDNSAddressConverted,
+		ExtraIFaceBlackList: extraIFaceBlackList,
 	}
 
 	if cmd.Flag(enableRosenpassFlag).Changed {
 		ic.RosenpassEnabled = &rosenpassEnabled
 	}
 
+	if cmd.Flag(rosenpassPermissiveFlag).Changed {
+		ic.RosenpassPermissive = &rosenpassPermissive
+	}
+
+	if cmd.Flag(serverSSHAllowedFlag).Changed {
+		ic.ServerSSHAllowed = &serverSSHAllowed
+	}
+
 	if cmd.Flag(interfaceNameFlag).Changed {
 		if err := parseInterfaceName(interfaceName); err != nil {
 			return err
@@ -106,10 +117,26 @@ 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
 	}
 
+	if cmd.Flag(disableAutoConnectFlag).Changed {
+		ic.DisableAutoConnect = &autoConnectDisabled
+
+		if autoConnectDisabled {
+			cmd.Println("Autoconnect has been disabled. The client won't connect automatically when the service starts.")
+		}
+
+		if !autoConnectDisabled {
+			cmd.Println("Autoconnect has been enabled. The client will connect automatically when the service starts.")
+		}
+	}
+
 	config, err := internal.UpdateOrCreateConfig(ic)
 	if err != nil {
 		return fmt.Errorf("get config file: %v", err)
@@ -125,11 +152,12 @@ 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()))
+
+	connectClient := internal.NewConnectClient(ctx, config, peer.NewRecorder(config.ManagementURL.String()))
+	return connectClient.Run()
 }
 
 func runInDaemonMode(ctx context.Context, cmd *cobra.Command) error {
-
 	customDNSAddressConverted, err := parseCustomDNSAddress(cmd.Flag(dnsResolverAddress).Changed)
 	if err != nil {
 		return err
@@ -170,6 +198,7 @@ func runInDaemonMode(ctx context.Context, cmd *cobra.Command) error {
 		CustomDNSAddress:     customDNSAddressConverted,
 		IsLinuxDesktopClient: isLinuxRunningDesktop(),
 		Hostname:             hostName,
+		ExtraIFaceBlacklist:  extraIFaceBlackList,
 	}
 
 	if rootCmd.PersistentFlags().Changed(preSharedKeyFlag) {
@@ -180,6 +209,18 @@ func runInDaemonMode(ctx context.Context, cmd *cobra.Command) error {
 		loginRequest.RosenpassEnabled = &rosenpassEnabled
 	}
 
+	if cmd.Flag(rosenpassPermissiveFlag).Changed {
+		loginRequest.RosenpassPermissive = &rosenpassPermissive
+	}
+
+	if cmd.Flag(serverSSHAllowedFlag).Changed {
+		loginRequest.ServerSSHAllowed = &serverSSHAllowed
+	}
+
+	if cmd.Flag(disableAutoConnectFlag).Changed {
+		loginRequest.DisableAutoConnect = &autoConnectDisabled
+	}
+
 	if cmd.Flag(interfaceNameFlag).Changed {
 		if err := parseInterfaceName(interfaceName); err != nil {
 			return err
@@ -192,6 +233,10 @@ func runInDaemonMode(ctx context.Context, cmd *cobra.Command) error {
 		loginRequest.WireguardPort = &wp
 	}
 
+	if cmd.Flag(networkMonitorFlag).Changed {
+		loginRequest.NetworkMonitor = &networkMonitor
+	}
+
 	var loginErr error
 
 	var loginResp *proto.LoginResponse

client/firewall/create_linux.go

@@ -42,20 +42,20 @@ func NewFirewall(context context.Context, iface IFaceMapper) (firewall.Manager,
 
 	switch check() {
 	case IPTABLES:
-		log.Debug("creating an iptables firewall manager")
+		log.Info("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")
+		log.Info("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")
+		log.Info("no firewall manager found, trying to use userspace packet filtering firewall")
 	}
 
 	if iface.IsUserspaceBind() {
@@ -85,16 +85,58 @@ func NewFirewall(context context.Context, iface IFaceMapper) (firewall.Manager,
 
 // 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
+	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
+		}
 	}
 
-	ip, err := iptables.NewWithProtocol(iptables.ProtocolIPv4)
-	if err != nil {
-		return UNKNOWN
+	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 isIptablesClientAvailable(ip) {
+
+	if useIPTABLES {
 		return IPTABLES
 	}
 

client/firewall/iptables/router_linux.go

@@ -87,12 +87,12 @@ func (i *routerManager) InsertRoutingRules(pair firewall.RouterPair) error {
 	return nil
 }
 
-// insertRoutingRule inserts an iptable rule
+// insertRoutingRule inserts an iptables rule
 func (i *routerManager) insertRoutingRule(keyFormat, table, chain, jump string, pair firewall.RouterPair) error {
 	var err error
 
 	ruleKey := firewall.GenKey(keyFormat, pair.ID)
-	rule := genRuleSpec(jump, ruleKey, pair.Source, pair.Destination)
+	rule := genRuleSpec(jump, pair.Source, pair.Destination)
 	existingRule, found := i.rules[ruleKey]
 	if found {
 		err = i.iptablesClient.DeleteIfExists(table, chain, existingRule...)
@@ -326,9 +326,9 @@ func (i *routerManager) createChain(table, newChain string) error {
 	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}
+// genRuleSpec generates rule specification
+func genRuleSpec(jump, source, destination string) []string {
+	return []string{"-s", source, "-d", destination, "-j", jump}
 }
 
 func getIptablesRuleType(table string) string {

client/firewall/iptables/router_linux_test.go

@@ -51,14 +51,12 @@ func TestIptablesManager_RestoreOrCreateContainers(t *testing.T) {
 		Destination: "100.100.100.0/24",
 		Masquerade:  true,
 	}
-	forward4RuleKey := firewall.GenKey(firewall.ForwardingFormat, pair.ID)
-	forward4Rule := genRuleSpec(routingFinalForwardJump, forward4RuleKey, pair.Source, pair.Destination)
+	forward4Rule := genRuleSpec(routingFinalForwardJump, 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)
+	nat4Rule := genRuleSpec(routingFinalNatJump, pair.Source, pair.Destination)
 
 	err = manager.iptablesClient.Insert(tableNat, chainRTNAT, 1, nat4Rule...)
 	require.NoError(t, err, "inserting rule should not return error")
@@ -92,7 +90,7 @@ func TestIptablesManager_InsertRoutingRules(t *testing.T) {
 			require.NoError(t, err, "forwarding pair should be inserted")
 
 			forwardRuleKey := firewall.GenKey(firewall.ForwardingFormat, testCase.InputPair.ID)
-			forwardRule := genRuleSpec(routingFinalForwardJump, forwardRuleKey, testCase.InputPair.Source, testCase.InputPair.Destination)
+			forwardRule := genRuleSpec(routingFinalForwardJump, 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)
@@ -103,7 +101,7 @@ func TestIptablesManager_InsertRoutingRules(t *testing.T) {
 			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)
+			inForwardRule := genRuleSpec(routingFinalForwardJump, 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)
@@ -114,7 +112,7 @@ func TestIptablesManager_InsertRoutingRules(t *testing.T) {
 			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)
+			natRule := genRuleSpec(routingFinalNatJump, 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)
@@ -130,7 +128,7 @@ func TestIptablesManager_InsertRoutingRules(t *testing.T) {
 			}
 
 			inNatRuleKey := firewall.GenKey(firewall.InNatFormat, testCase.InputPair.ID)
-			inNatRule := genRuleSpec(routingFinalNatJump, inNatRuleKey, firewall.GetInPair(testCase.InputPair).Source, firewall.GetInPair(testCase.InputPair).Destination)
+			inNatRule := genRuleSpec(routingFinalNatJump, 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)
@@ -167,25 +165,25 @@ func TestIptablesManager_RemoveRoutingRules(t *testing.T) {
 			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)
+			forwardRule := genRuleSpec(routingFinalForwardJump, testCase.InputPair.Source, testCase.InputPair.Destination)
 
 			err = iptablesClient.Insert(tableFilter, chainRTFWD, 1, forwardRule...)
 			require.NoError(t, err, "inserting rule should not return error")
 
 			inForwardRuleKey := firewall.GenKey(firewall.InForwardingFormat, testCase.InputPair.ID)
-			inForwardRule := genRuleSpec(routingFinalForwardJump, inForwardRuleKey, firewall.GetInPair(testCase.InputPair).Source, firewall.GetInPair(testCase.InputPair).Destination)
+			inForwardRule := genRuleSpec(routingFinalForwardJump, firewall.GetInPair(testCase.InputPair).Source, firewall.GetInPair(testCase.InputPair).Destination)
 
 			err = iptablesClient.Insert(tableFilter, chainRTFWD, 1, inForwardRule...)
 			require.NoError(t, err, "inserting rule should not return error")
 
 			natRuleKey := firewall.GenKey(firewall.NatFormat, testCase.InputPair.ID)
-			natRule := genRuleSpec(routingFinalNatJump, natRuleKey, testCase.InputPair.Source, testCase.InputPair.Destination)
+			natRule := genRuleSpec(routingFinalNatJump, 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)
+			inNatRule := genRuleSpec(routingFinalNatJump, 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")

client/firewall/uspfilter/allow_netbird_windows.go

@@ -64,15 +64,18 @@ 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 +90,23 @@ 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/internal/auth/oauth.go

@@ -26,7 +26,7 @@ type HTTPClient interface {
 }
 
 // AuthFlowInfo holds information for the OAuth 2.0  authorization flow
-type AuthFlowInfo struct {
+type AuthFlowInfo struct { //nolint:revive
 	DeviceCode              string `json:"device_code"`
 	UserCode                string `json:"user_code"`
 	VerificationURI         string `json:"verification_uri"`

client/internal/config.go

@@ -5,6 +5,8 @@ import (
 	"fmt"
 	"net/url"
 	"os"
+	"reflect"
+	"strings"
 
 	log "github.com/sirupsen/logrus"
 	"golang.zx2c4.com/wireguard/wgctrl/wgtypes"
@@ -30,20 +32,27 @@ const (
 	DefaultAdminURL = "https://app.netbird.io:443"
 )
 
-var defaultInterfaceBlacklist = []string{iface.WgInterfaceDefault, "wt", "utun", "tun0", "zt", "ZeroTier", "wg", "ts",
-	"Tailscale", "tailscale", "docker", "veth", "br-", "lo"}
+var defaultInterfaceBlacklist = []string{
+	iface.WgInterfaceDefault, "wt", "utun", "tun0", "zt", "ZeroTier", "wg", "ts",
+	"Tailscale", "tailscale", "docker", "veth", "br-", "lo",
+}
 
 // ConfigInput carries configuration changes to the client
 type ConfigInput struct {
-	ManagementURL    string
-	AdminURL         string
-	ConfigPath       string
-	PreSharedKey     *string
-	NATExternalIPs   []string
-	CustomDNSAddress []byte
-	RosenpassEnabled *bool
-	InterfaceName    *string
-	WireguardPort    *int
+	ManagementURL       string
+	AdminURL            string
+	ConfigPath          string
+	PreSharedKey        *string
+	ServerSSHAllowed    *bool
+	NATExternalIPs      []string
+	CustomDNSAddress    []byte
+	RosenpassEnabled    *bool
+	RosenpassPermissive *bool
+	InterfaceName       *string
+	WireguardPort       *int
+	NetworkMonitor      *bool
+	DisableAutoConnect  *bool
+	ExtraIFaceBlackList []string
 }
 
 // Config Configuration type
@@ -55,9 +64,12 @@ type Config struct {
 	AdminURL             *url.URL
 	WgIface              string
 	WgPort               int
+	NetworkMonitor       bool
 	IFaceBlackList       []string
 	DisableIPv6Discovery bool
 	RosenpassEnabled     bool
+	RosenpassPermissive  bool
+	ServerSSHAllowed     *bool
 	// SSHKey is a private SSH key in a PEM format
 	SSHKey string
 
@@ -79,6 +91,10 @@ type Config struct {
 	NATExternalIPs []string
 	// CustomDNSAddress sets the DNS resolver listening address in format ip:port
 	CustomDNSAddress string
+
+	// DisableAutoConnect determines whether the client should not start with the service
+	// it's set to false by default due to backwards compatibility
+	DisableAutoConnect bool
 }
 
 // ReadConfig read config file and return with Config. If it is not exists create a new with default values
@@ -88,6 +104,15 @@ func ReadConfig(configPath string) (*Config, error) {
 		if _, err := util.ReadJson(configPath, config); err != nil {
 			return nil, err
 		}
+		// initialize through apply() without changes
+		if changed, err := config.apply(ConfigInput{}); err != nil {
+			return nil, err
+		} else if changed {
+			if err = WriteOutConfig(configPath, config); err != nil {
+				return nil, err
+			}
+		}
+
 		return config, nil
 	}
 
@@ -139,68 +164,15 @@ func WriteOutConfig(path string, config *Config) error {
 
 // createNewConfig creates a new config generating a new Wireguard key and saving to file
 func createNewConfig(input ConfigInput) (*Config, error) {
-	wgKey := generateKey()
-	pem, err := ssh.GeneratePrivateKey(ssh.ED25519)
-	if err != nil {
-		return nil, err
-	}
-
 	config := &Config{
-		SSHKey:               string(pem),
-		PrivateKey:           wgKey,
-		IFaceBlackList:       []string{},
-		DisableIPv6Discovery: false,
-		NATExternalIPs:       input.NATExternalIPs,
-		CustomDNSAddress:     string(input.CustomDNSAddress),
+		// defaults to false only for new (post 0.26) configurations
+		ServerSSHAllowed: util.False(),
 	}
 
-	defaultManagementURL, err := parseURL("Management URL", DefaultManagementURL)
-	if err != nil {
+	if _, err := config.apply(input); err != nil {
 		return nil, err
 	}
 
-	config.ManagementURL = defaultManagementURL
-	if input.ManagementURL != "" {
-		URL, err := parseURL("Management URL", input.ManagementURL)
-		if err != nil {
-			return nil, err
-		}
-		config.ManagementURL = URL
-	}
-
-	config.WgPort = iface.DefaultWgPort
-	if input.WireguardPort != nil {
-		config.WgPort = *input.WireguardPort
-	}
-
-	config.WgIface = iface.WgInterfaceDefault
-	if input.InterfaceName != nil {
-		config.WgIface = *input.InterfaceName
-	}
-
-	if input.PreSharedKey != nil {
-		config.PreSharedKey = *input.PreSharedKey
-	}
-
-	if input.RosenpassEnabled != nil {
-		config.RosenpassEnabled = *input.RosenpassEnabled
-	}
-
-	defaultAdminURL, err := parseURL("Admin URL", DefaultAdminURL)
-	if err != nil {
-		return nil, err
-	}
-
-	config.AdminURL = defaultAdminURL
-	if input.AdminURL != "" {
-		newURL, err := parseURL("Admin Panel URL", input.AdminURL)
-		if err != nil {
-			return nil, err
-		}
-		config.AdminURL = newURL
-	}
-
-	config.IFaceBlackList = defaultInterfaceBlacklist
 	return config, nil
 }
 
@@ -211,77 +183,12 @@ func update(input ConfigInput) (*Config, error) {
 		return nil, err
 	}
 
-	refresh := false
-
-	if input.ManagementURL != "" && config.ManagementURL.String() != input.ManagementURL {
-		log.Infof("new Management URL provided, updated to %s (old value %s)",
-			input.ManagementURL, config.ManagementURL)
-		newURL, err := parseURL("Management URL", input.ManagementURL)
-		if err != nil {
-			return nil, err
-		}
-		config.ManagementURL = newURL
-		refresh = true
+	updated, err := config.apply(input)
+	if err != nil {
+		return nil, err
 	}
 
-	if input.AdminURL != "" && (config.AdminURL == nil || config.AdminURL.String() != input.AdminURL) {
-		log.Infof("new Admin Panel URL provided, updated to %s (old value %s)",
-			input.AdminURL, config.AdminURL)
-		newURL, err := parseURL("Admin Panel URL", input.AdminURL)
-		if err != nil {
-			return nil, err
-		}
-		config.AdminURL = newURL
-		refresh = true
-	}
-
-	if input.PreSharedKey != nil && config.PreSharedKey != *input.PreSharedKey {
-		log.Infof("new pre-shared key provided, replacing old key")
-		config.PreSharedKey = *input.PreSharedKey
-		refresh = true
-	}
-
-	if config.SSHKey == "" {
-		pem, err := ssh.GeneratePrivateKey(ssh.ED25519)
-		if err != nil {
-			return nil, err
-		}
-		config.SSHKey = string(pem)
-		refresh = true
-	}
-
-	if config.WgPort == 0 {
-		config.WgPort = iface.DefaultWgPort
-		refresh = true
-	}
-
-	if input.WireguardPort != nil {
-		config.WgPort = *input.WireguardPort
-		refresh = true
-	}
-
-	if input.InterfaceName != nil {
-		config.WgIface = *input.InterfaceName
-		refresh = true
-	}
-
-	if input.NATExternalIPs != nil && len(config.NATExternalIPs) != len(input.NATExternalIPs) {
-		config.NATExternalIPs = input.NATExternalIPs
-		refresh = true
-	}
-
-	if input.CustomDNSAddress != nil {
-		config.CustomDNSAddress = string(input.CustomDNSAddress)
-		refresh = true
-	}
-
-	if input.RosenpassEnabled != nil {
-		config.RosenpassEnabled = *input.RosenpassEnabled
-		refresh = true
-	}
-
-	if refresh {
-		// since we have new management URL, we need to update config file
+	if updated {
 		if err := util.WriteJson(input.ConfigPath, config); err != nil {
 			return nil, err
 		}
@@ -290,6 +197,169 @@ func update(input ConfigInput) (*Config, error) {
 	return config, nil
 }
 
+func (config *Config) apply(input ConfigInput) (updated bool, err error) {
+	if config.ManagementURL == nil {
+		log.Infof("using default Management URL %s", DefaultManagementURL)
+		config.ManagementURL, err = parseURL("Management URL", DefaultManagementURL)
+		if err != nil {
+			return false, err
+		}
+	}
+	if input.ManagementURL != "" && input.ManagementURL != config.ManagementURL.String() {
+		log.Infof("new Management URL provided, updated to %#v (old value %#v)",
+			input.ManagementURL, config.ManagementURL.String())
+		URL, err := parseURL("Management URL", input.ManagementURL)
+		if err != nil {
+			return false, err
+		}
+		config.ManagementURL = URL
+		updated = true
+	} else if config.ManagementURL == nil {
+		log.Infof("using default Management URL %s", DefaultManagementURL)
+		config.ManagementURL, err = parseURL("Management URL", DefaultManagementURL)
+		if err != nil {
+			return false, err
+		}
+	}
+
+	if config.AdminURL == nil {
+		log.Infof("using default Admin URL %s", DefaultManagementURL)
+		config.AdminURL, err = parseURL("Admin URL", DefaultAdminURL)
+		if err != nil {
+			return false, err
+		}
+	}
+	if input.AdminURL != "" && input.AdminURL != config.AdminURL.String() {
+		log.Infof("new Admin Panel URL provided, updated to %#v (old value %#v)",
+			input.AdminURL, config.AdminURL.String())
+		newURL, err := parseURL("Admin Panel URL", input.AdminURL)
+		if err != nil {
+			return updated, err
+		}
+		config.AdminURL = newURL
+		updated = true
+	}
+
+	if config.PrivateKey == "" {
+		log.Infof("generated new Wireguard key")
+		config.PrivateKey = generateKey()
+		updated = true
+	}
+
+	if config.SSHKey == "" {
+		log.Infof("generated new SSH key")
+		pem, err := ssh.GeneratePrivateKey(ssh.ED25519)
+		if err != nil {
+			return false, err
+		}
+		config.SSHKey = string(pem)
+		updated = true
+	}
+
+	if input.WireguardPort != nil && *input.WireguardPort != config.WgPort {
+		log.Infof("updating Wireguard port %d (old value %d)",
+			*input.WireguardPort, config.WgPort)
+		config.WgPort = *input.WireguardPort
+		updated = true
+	} else if config.WgPort == 0 {
+		config.WgPort = iface.DefaultWgPort
+		log.Infof("using default Wireguard port %d", config.WgPort)
+		updated = true
+	}
+
+	if input.InterfaceName != nil && *input.InterfaceName != config.WgIface {
+		log.Infof("updating Wireguard interface %#v (old value %#v)",
+			*input.InterfaceName, config.WgIface)
+		config.WgIface = *input.InterfaceName
+		updated = true
+	} else if config.WgIface == "" {
+		config.WgIface = iface.WgInterfaceDefault
+		log.Infof("using default Wireguard interface %s", config.WgIface)
+		updated = true
+	}
+
+	if input.NATExternalIPs != nil && !reflect.DeepEqual(config.NATExternalIPs, input.NATExternalIPs) {
+		log.Infof("updating NAT External IP [ %s ] (old value: [ %s ])",
+			strings.Join(input.NATExternalIPs, " "),
+			strings.Join(config.NATExternalIPs, " "))
+		config.NATExternalIPs = input.NATExternalIPs
+		updated = true
+	}
+
+	if input.PreSharedKey != nil && *input.PreSharedKey != config.PreSharedKey {
+		log.Infof("new pre-shared key provided, replacing old key")
+		config.PreSharedKey = *input.PreSharedKey
+		updated = true
+	}
+
+	if input.RosenpassEnabled != nil && *input.RosenpassEnabled != config.RosenpassEnabled {
+		log.Infof("switching Rosenpass to %t", *input.RosenpassEnabled)
+		config.RosenpassEnabled = *input.RosenpassEnabled
+		updated = true
+	}
+
+	if input.RosenpassPermissive != nil && *input.RosenpassPermissive != config.RosenpassPermissive {
+		log.Infof("switching Rosenpass permissive to %t", *input.RosenpassPermissive)
+		config.RosenpassPermissive = *input.RosenpassPermissive
+		updated = true
+	}
+
+	if input.NetworkMonitor != nil && *input.NetworkMonitor != config.NetworkMonitor {
+		log.Infof("switching Network Monitor to %t", *input.NetworkMonitor)
+		config.NetworkMonitor = *input.NetworkMonitor
+		updated = true
+	}
+
+	if input.CustomDNSAddress != nil && string(input.CustomDNSAddress) != config.CustomDNSAddress {
+		log.Infof("updating custom DNS address %#v (old value %#v)",
+			string(input.CustomDNSAddress), config.CustomDNSAddress)
+		config.CustomDNSAddress = string(input.CustomDNSAddress)
+		updated = true
+	}
+
+	if len(config.IFaceBlackList) == 0 {
+		log.Infof("filling in interface blacklist with defaults: [ %s ]",
+			strings.Join(defaultInterfaceBlacklist, " "))
+		config.IFaceBlackList = append(config.IFaceBlackList, defaultInterfaceBlacklist...)
+		updated = true
+	}
+
+	if len(input.ExtraIFaceBlackList) > 0 {
+		for _, iFace := range util.SliceDiff(input.ExtraIFaceBlackList, config.IFaceBlackList) {
+			log.Infof("adding new entry to interface blacklist: %s", iFace)
+			config.IFaceBlackList = append(config.IFaceBlackList, iFace)
+			updated = true
+		}
+	}
+
+	if input.DisableAutoConnect != nil && *input.DisableAutoConnect != config.DisableAutoConnect {
+		if *input.DisableAutoConnect {
+			log.Infof("turning off automatic connection on startup")
+		} else {
+			log.Infof("enabling automatic connection on startup")
+		}
+		config.DisableAutoConnect = *input.DisableAutoConnect
+		updated = true
+	}
+
+	if input.ServerSSHAllowed != nil && *input.ServerSSHAllowed != *config.ServerSSHAllowed {
+		if *input.ServerSSHAllowed {
+			log.Infof("enabling SSH server")
+		} else {
+			log.Infof("disabling SSH server")
+		}
+		config.ServerSSHAllowed = input.ServerSSHAllowed
+		updated = true
+	} else if config.ServerSSHAllowed == nil {
+		// enables SSH for configs from old versions to preserve backwards compatibility
+		log.Infof("falling back to enabled SSH server for pre-existing configuration")
+		config.ServerSSHAllowed = util.True()
+		updated = true
+	}
+
+	return updated, nil
+}
+
 // parseURL parses and validates a service URL
 func parseURL(serviceName, serviceURL string) (*url.URL, error) {
 	parsedMgmtURL, err := url.ParseRequestURI(serviceURL)
@@ -344,7 +414,6 @@ func configFileIsExists(path string) bool {
 // If it can switch, then it updates the config and returns a new one. Otherwise, it returns the provided config.
 // The check is performed only for the NetBird's managed version.
 func UpdateOldManagementURL(ctx context.Context, config *Config, configPath string) (*Config, error) {
-
 	defaultManagementURL, err := parseURL("Management URL", DefaultManagementURL)
 	if err != nil {
 		return nil, err

client/internal/config_test.go

@@ -18,7 +18,6 @@ func TestGetConfig(t *testing.T) {
 	config, err := UpdateOrCreateConfig(ConfigInput{
 		ConfigPath: filepath.Join(t.TempDir(), "config.json"),
 	})
-
 	if err != nil {
 		return
 	}
@@ -86,6 +85,26 @@ func TestGetConfig(t *testing.T) {
 	assert.Equal(t, readConf.(*Config).ManagementURL.String(), newManagementURL)
 }
 
+func TestExtraIFaceBlackList(t *testing.T) {
+	extraIFaceBlackList := []string{"eth1"}
+	path := filepath.Join(t.TempDir(), "config.json")
+	config, err := UpdateOrCreateConfig(ConfigInput{
+		ConfigPath:          path,
+		ExtraIFaceBlackList: extraIFaceBlackList,
+	})
+	if err != nil {
+		return
+	}
+
+	assert.Contains(t, config.IFaceBlackList, "eth1")
+	readConf, err := util.ReadJson(path, config)
+	if err != nil {
+		return
+	}
+
+	assert.Contains(t, readConf.(*Config).IFaceBlackList, "eth1")
+}
+
 func TestHiddenPreSharedKey(t *testing.T) {
 	hidden := "**********"
 	samplePreSharedKey := "mysecretpresharedkey"
@@ -111,7 +130,6 @@ func TestHiddenPreSharedKey(t *testing.T) {
 				ConfigPath:   cfgFile,
 				PreSharedKey: tt.preSharedKey,
 			})
-
 			if err != nil {
 				t.Fatalf("failed to get cfg: %s", err)
 			}

client/internal/connect.go

@@ -2,8 +2,13 @@ package internal
 
 import (
 	"context"
+	"errors"
 	"fmt"
+	"net"
+	"runtime"
+	"runtime/debug"
 	"strings"
+	"sync"
 	"time"
 
 	"github.com/cenkalti/backoff/v4"
@@ -22,32 +27,49 @@ import (
 	mgm "github.com/netbirdio/netbird/management/client"
 	mgmProto "github.com/netbirdio/netbird/management/proto"
 	signal "github.com/netbirdio/netbird/signal/client"
+	"github.com/netbirdio/netbird/util"
 	"github.com/netbirdio/netbird/version"
 )
 
-// RunClient with main logic.
-func RunClient(ctx context.Context, config *Config, statusRecorder *peer.Status) error {
-	return runClient(ctx, config, statusRecorder, MobileDependency{}, nil, nil, nil, nil)
+type ConnectClient struct {
+	ctx            context.Context
+	config         *Config
+	statusRecorder *peer.Status
+	engine         *Engine
+	engineMutex    sync.Mutex
 }
 
-// RunClientWithProbes runs the client's main logic with probes attached
-func RunClientWithProbes(
+func NewConnectClient(
 	ctx context.Context,
 	config *Config,
 	statusRecorder *peer.Status,
+
+) *ConnectClient {
+	return &ConnectClient{
+		ctx:            ctx,
+		config:         config,
+		statusRecorder: statusRecorder,
+		engineMutex:    sync.Mutex{},
+	}
+}
+
+// Run with main logic.
+func (c *ConnectClient) Run() error {
+	return c.run(MobileDependency{}, nil, nil, nil, nil)
+}
+
+// RunWithProbes runs the client's main logic with probes attached
+func (c *ConnectClient) RunWithProbes(
 	mgmProbe *Probe,
 	signalProbe *Probe,
 	relayProbe *Probe,
 	wgProbe *Probe,
 ) error {
-	return runClient(ctx, config, statusRecorder, MobileDependency{}, mgmProbe, signalProbe, relayProbe, wgProbe)
+	return c.run(MobileDependency{}, mgmProbe, signalProbe, relayProbe, wgProbe)
 }
 
-// RunClientMobile with main logic on mobile system
-func RunClientMobile(
-	ctx context.Context,
-	config *Config,
-	statusRecorder *peer.Status,
+// RunOnAndroid with main logic on mobile system
+func (c *ConnectClient) RunOnAndroid(
 	tunAdapter iface.TunAdapter,
 	iFaceDiscover stdnet.ExternalIFaceDiscover,
 	networkChangeListener listener.NetworkChangeListener,
@@ -62,36 +84,45 @@ func RunClientMobile(
 		HostDNSAddresses:      dnsAddresses,
 		DnsReadyListener:      dnsReadyListener,
 	}
-	return runClient(ctx, config, statusRecorder, mobileDependency, nil, nil, nil, nil)
+	return c.run(mobileDependency, nil, nil, nil, nil)
 }
 
-func RunClientiOS(
-	ctx context.Context,
-	config *Config,
-	statusRecorder *peer.Status,
+func (c *ConnectClient) RunOniOS(
 	fileDescriptor int32,
 	networkChangeListener listener.NetworkChangeListener,
 	dnsManager dns.IosDnsManager,
 ) error {
+	// Set GC percent to 5% to reduce memory usage as iOS only allows 50MB of memory for the extension.
+	debug.SetGCPercent(5)
+
 	mobileDependency := MobileDependency{
 		FileDescriptor:        fileDescriptor,
 		NetworkChangeListener: networkChangeListener,
 		DnsManager:            dnsManager,
 	}
-	return runClient(ctx, config, statusRecorder, mobileDependency, nil, nil, nil, nil)
+	return c.run(mobileDependency, nil, nil, nil, nil)
 }
 
-func runClient(
-	ctx context.Context,
-	config *Config,
-	statusRecorder *peer.Status,
+func (c *ConnectClient) run(
 	mobileDependency MobileDependency,
 	mgmProbe *Probe,
 	signalProbe *Probe,
 	relayProbe *Probe,
 	wgProbe *Probe,
 ) error {
-	log.Infof("starting NetBird client version %s", version.NetbirdVersion())
+	defer func() {
+		if r := recover(); r != nil {
+			log.Panicf("Panic occurred: %v, stack trace: %s", r, string(debug.Stack()))
+		}
+	}()
+
+	log.Infof("starting NetBird client version %s on %s/%s", version.NetbirdVersion(), runtime.GOOS, runtime.GOARCH)
+
+	// Check if client was not shut down in a clean way and restore DNS config if required.
+	// Otherwise, we might not be able to connect to the management server to retrieve new config.
+	if err := dns.CheckUncleanShutdown(c.config.WgIface); err != nil {
+		log.Errorf("checking unclean shutdown error: %s", err)
+	}
 
 	backOff := &backoff.ExponentialBackOff{
 		InitialInterval:     time.Second,
@@ -103,7 +134,7 @@ func runClient(
 		Clock:               backoff.SystemClock,
 	}
 
-	state := CtxGetState(ctx)
+	state := CtxGetState(c.ctx)
 	defer func() {
 		s, err := state.Status()
 		if err != nil || s != StatusNeedsLogin {
@@ -112,49 +143,49 @@ func runClient(
 	}()
 
 	wrapErr := state.Wrap
-	myPrivateKey, err := wgtypes.ParseKey(config.PrivateKey)
+	myPrivateKey, err := wgtypes.ParseKey(c.config.PrivateKey)
 	if err != nil {
-		log.Errorf("failed parsing Wireguard key %s: [%s]", config.PrivateKey, err.Error())
+		log.Errorf("failed parsing Wireguard key %s: [%s]", c.config.PrivateKey, err.Error())
 		return wrapErr(err)
 	}
 
 	var mgmTlsEnabled bool
-	if config.ManagementURL.Scheme == "https" {
+	if c.config.ManagementURL.Scheme == "https" {
 		mgmTlsEnabled = true
 	}
 
-	publicSSHKey, err := ssh.GeneratePublicKey([]byte(config.SSHKey))
+	publicSSHKey, err := ssh.GeneratePublicKey([]byte(c.config.SSHKey))
 	if err != nil {
 		return err
 	}
 
-	defer statusRecorder.ClientStop()
+	defer c.statusRecorder.ClientStop()
 	operation := func() error {
 		// if context cancelled we not start new backoff cycle
 		select {
-		case <-ctx.Done():
+		case <-c.ctx.Done():
 			return nil
 		default:
 		}
 
 		state.Set(StatusConnecting)
 
-		engineCtx, cancel := context.WithCancel(ctx)
+		engineCtx, cancel := context.WithCancel(c.ctx)
 		defer func() {
-			statusRecorder.MarkManagementDisconnected(state.err)
-			statusRecorder.CleanLocalPeerState()
+			c.statusRecorder.MarkManagementDisconnected(state.err)
+			c.statusRecorder.CleanLocalPeerState()
 			cancel()
 		}()
 
-		log.Debugf("connecting to the Management service %s", config.ManagementURL.Host)
-		mgmClient, err := mgm.NewClient(engineCtx, config.ManagementURL.Host, myPrivateKey, mgmTlsEnabled)
+		log.Debugf("connecting to the Management service %s", c.config.ManagementURL.Host)
+		mgmClient, err := mgm.NewClient(engineCtx, c.config.ManagementURL.Host, myPrivateKey, mgmTlsEnabled)
 		if err != nil {
 			return wrapErr(gstatus.Errorf(codes.FailedPrecondition, "failed connecting to Management Service : %s", err))
 		}
-		mgmNotifier := statusRecorderToMgmConnStateNotifier(statusRecorder)
+		mgmNotifier := statusRecorderToMgmConnStateNotifier(c.statusRecorder)
 		mgmClient.SetConnStateListener(mgmNotifier)
 
-		log.Debugf("connected to the Management service %s", config.ManagementURL.Host)
+		log.Debugf("connected to the Management service %s", c.config.ManagementURL.Host)
 		defer func() {
 			err = mgmClient.Close()
 			if err != nil {
@@ -172,7 +203,7 @@ func runClient(
 			}
 			return wrapErr(err)
 		}
-		statusRecorder.MarkManagementConnected()
+		c.statusRecorder.MarkManagementConnected()
 
 		localPeerState := peer.LocalPeerState{
 			IP:              loginResp.GetPeerConfig().GetAddress(),
@@ -181,18 +212,18 @@ func runClient(
 			FQDN:            loginResp.GetPeerConfig().GetFqdn(),
 		}
 
-		statusRecorder.UpdateLocalPeerState(localPeerState)
+		c.statusRecorder.UpdateLocalPeerState(localPeerState)
 
 		signalURL := fmt.Sprintf("%s://%s",
 			strings.ToLower(loginResp.GetWiretrusteeConfig().GetSignal().GetProtocol().String()),
 			loginResp.GetWiretrusteeConfig().GetSignal().GetUri(),
 		)
 
-		statusRecorder.UpdateSignalAddress(signalURL)
+		c.statusRecorder.UpdateSignalAddress(signalURL)
 
-		statusRecorder.MarkSignalDisconnected(nil)
+		c.statusRecorder.MarkSignalDisconnected(nil)
 		defer func() {
-			statusRecorder.MarkSignalDisconnected(state.err)
+			c.statusRecorder.MarkSignalDisconnected(state.err)
 		}()
 
 		// with the global Wiretrustee config in hand connect (just a connection, no stream yet) Signal
@@ -208,35 +239,38 @@ func runClient(
 			}
 		}()
 
-		signalNotifier := statusRecorderToSignalConnStateNotifier(statusRecorder)
+		signalNotifier := statusRecorderToSignalConnStateNotifier(c.statusRecorder)
 		signalClient.SetConnStateListener(signalNotifier)
 
-		statusRecorder.MarkSignalConnected()
+		c.statusRecorder.MarkSignalConnected()
 
 		peerConfig := loginResp.GetPeerConfig()
 
-		engineConfig, err := createEngineConfig(myPrivateKey, config, peerConfig)
+		engineConfig, err := createEngineConfig(myPrivateKey, c.config, peerConfig)
 		if err != nil {
 			log.Error(err)
 			return wrapErr(err)
 		}
 
-		engine := NewEngineWithProbes(engineCtx, cancel, signalClient, mgmClient, engineConfig, mobileDependency, statusRecorder, mgmProbe, signalProbe, relayProbe, wgProbe)
-		err = engine.Start()
+		c.engineMutex.Lock()
+		c.engine = NewEngineWithProbes(engineCtx, cancel, signalClient, mgmClient, engineConfig, mobileDependency, c.statusRecorder, mgmProbe, signalProbe, relayProbe, wgProbe)
+		c.engineMutex.Unlock()
+
+		err = c.engine.Start()
 		if err != nil {
 			log.Errorf("error while starting Netbird Connection Engine: %s", err)
 			return wrapErr(err)
 		}
 
-		log.Print("Netbird engine started, my IP is: ", peerConfig.Address)
+		log.Infof("Netbird engine started, the IP is: %s", peerConfig.GetAddress())
 		state.Set(StatusConnected)
 
 		<-engineCtx.Done()
-		statusRecorder.ClientTeardown()
+		c.statusRecorder.ClientTeardown()
 
 		backOff.Reset()
 
-		err = engine.Stop()
+		err = c.engine.Stop()
 		if err != nil {
 			log.Errorf("failed stopping engine %v", err)
 			return wrapErr(err)
@@ -244,14 +278,14 @@ func runClient(
 
 		log.Info("stopped NetBird client")
 
-		if _, err := state.Status(); err == ErrResetConnection {
+		if _, err := state.Status(); errors.Is(err, ErrResetConnection) {
 			return err
 		}
 
 		return nil
 	}
 
-	statusRecorder.ClientStart()
+	c.statusRecorder.ClientStart()
 	err = backoff.Retry(operation, backOff)
 	if err != nil {
 		log.Debugf("exiting client retry loop due to unrecoverable error: %s", err)
@@ -263,6 +297,14 @@ func runClient(
 	return nil
 }
 
+func (c *ConnectClient) Engine() *Engine {
+	var e *Engine
+	c.engineMutex.Lock()
+	e = c.engine
+	c.engineMutex.Unlock()
+	return e
+}
+
 // createEngineConfig converts configuration received from Management Service to EngineConfig
 func createEngineConfig(key wgtypes.Key, config *Config, peerConfig *mgmProto.PeerConfig) (*EngineConfig, error) {
 	engineConf := &EngineConfig{
@@ -272,10 +314,13 @@ func createEngineConfig(key wgtypes.Key, config *Config, peerConfig *mgmProto.Pe
 		DisableIPv6Discovery: config.DisableIPv6Discovery,
 		WgPrivateKey:         key,
 		WgPort:               config.WgPort,
+		NetworkMonitor:       config.NetworkMonitor,
 		SSHKey:               []byte(config.SSHKey),
 		NATExternalIPs:       config.NATExternalIPs,
 		CustomDNSAddress:     config.CustomDNSAddress,
 		RosenpassEnabled:     config.RosenpassEnabled,
+		RosenpassPermissive:  config.RosenpassPermissive,
+		ServerSSHAllowed:     util.ReturnBoolWithDefaultTrue(config.ServerSSHAllowed),
 	}
 
 	if config.PreSharedKey != "" {
@@ -286,6 +331,15 @@ func createEngineConfig(key wgtypes.Key, config *Config, peerConfig *mgmProto.Pe
 		engineConf.PreSharedKey = &preSharedKey
 	}
 
+	port, err := freePort(config.WgPort)
+	if err != nil {
+		return nil, err
+	}
+	if port != config.WgPort {
+		log.Infof("using %d as wireguard port: %d is in use", port, config.WgPort)
+	}
+	engineConf.WgPort = port
+
 	return engineConf, nil
 }
 
@@ -335,3 +389,20 @@ func statusRecorderToSignalConnStateNotifier(statusRecorder *peer.Status) signal
 	notifier, _ := sri.(signal.ConnStateNotifier)
 	return notifier
 }
+
+func freePort(start int) (int, error) {
+	addr := net.UDPAddr{}
+	if start == 0 {
+		start = iface.DefaultWgPort
+	}
+	for x := start; x <= 65535; x++ {
+		addr.Port = x
+		conn, err := net.ListenUDP("udp", &addr)
+		if err != nil {
+			continue
+		}
+		conn.Close()
+		return x, nil
+	}
+	return 0, errors.New("no free ports")
+}

client/internal/connect_test.go

@@ -0,0 +1,57 @@
+package internal
+
+import (
+	"net"
+	"testing"
+)
+
+func Test_freePort(t *testing.T) {
+	tests := []struct {
+		name    string
+		port    int
+		want    int
+		wantErr bool
+	}{
+		{
+			name:    "available",
+			port:    51820,
+			want:    51820,
+			wantErr: false,
+		},
+		{
+			name:    "notavailable",
+			port:    51830,
+			want:    51831,
+			wantErr: false,
+		},
+		{
+			name:    "noports",
+			port:    65535,
+			want:    0,
+			wantErr: true,
+		},
+	}
+	for _, tt := range tests {
+
+		c1, err := net.ListenUDP("udp", &net.UDPAddr{Port: 51830})
+		if err != nil {
+			t.Errorf("freePort error = %v", err)
+		}
+		c2, err := net.ListenUDP("udp", &net.UDPAddr{Port: 65535})
+		if err != nil {
+			t.Errorf("freePort error = %v", err)
+		}
+		t.Run(tt.name, func(t *testing.T) {
+			got, err := freePort(tt.port)
+			if (err != nil) != tt.wantErr {
+				t.Errorf("freePort() error = %v, wantErr %v", err, tt.wantErr)
+				return
+			}
+			if got != tt.want {
+				t.Errorf("freePort() = %v, want %v", got, tt.want)
+			}
+		})
+		c1.Close()
+		c2.Close()
+	}
+}

client/internal/dns/dbus_linux.go

@@ -4,9 +4,11 @@ package dns
 
 import (
 	"context"
+	"fmt"
+	"time"
+
 	"github.com/godbus/dbus/v5"
 	log "github.com/sirupsen/logrus"
-	"time"
 )
 
 const dbusDefaultFlag = 0
@@ -14,6 +16,7 @@ const dbusDefaultFlag = 0
 func isDbusListenerRunning(dest string, path dbus.ObjectPath) bool {
 	obj, closeConn, err := getDbusObject(dest, path)
 	if err != nil {
+		log.Tracef("error getting dbus object: %s", err)
 		return false
 	}
 	defer closeConn()
@@ -21,14 +24,18 @@ func isDbusListenerRunning(dest string, path dbus.ObjectPath) bool {
 	ctx, cancel := context.WithTimeout(context.TODO(), 5*time.Second)
 	defer cancel()
 
-	err = obj.CallWithContext(ctx, "org.freedesktop.DBus.Peer.Ping", 0).Store()
-	return err == nil
+	if err = obj.CallWithContext(ctx, "org.freedesktop.DBus.Peer.Ping", 0).Store(); err != nil {
+		log.Tracef("error calling dbus: %s", err)
+		return false
+	}
+
+	return true
 }
 
 func getDbusObject(dest string, path dbus.ObjectPath) (dbus.BusObject, func(), error) {
 	conn, err := dbus.SystemBus()
 	if err != nil {
-		return nil, nil, err
+		return nil, nil, fmt.Errorf("get dbus: %w", err)
 	}
 	obj := conn.Object(dest, path)
 

client/internal/dns/file_linux.go

@@ -5,8 +5,10 @@ package dns
 import (
 	"bytes"
 	"fmt"
+	"net/netip"
 	"os"
 	"strings"
+	"time"
 
 	log "github.com/sirupsen/logrus"
 )
@@ -22,10 +24,16 @@ const (
 	fileMaxNumberOfSearchDomains = 6
 )
 
+const (
+	dnsFailoverTimeout  = 4 * time.Second
+	dnsFailoverAttempts = 1
+)
+
 type fileConfigurator struct {
 	repair *repair
 
-	originalPerms os.FileMode
+	originalPerms  os.FileMode
+	nbNameserverIP string
 }
 
 func newFileConfigurator() (hostManager, error) {
@@ -39,44 +47,40 @@ func (f *fileConfigurator) supportCustomPort() bool {
 }
 
 func (f *fileConfigurator) applyDNSConfig(config HostDNSConfig) error {
-	backupFileExist := false
-	_, err := os.Stat(fileDefaultResolvConfBackupLocation)
-	if err == nil {
-		backupFileExist = true
-	}
-
+	backupFileExist := f.isBackupFileExist()
 	if !config.RouteAll {
 		if backupFileExist {
-			err = f.restore()
+			f.repair.stopWatchFileChanges()
+			err := f.restore()
 			if err != nil {
-				return fmt.Errorf("unable to configure DNS for this peer using file manager without a Primary nameserver group. Restoring the original file return err: %s", err)
+				return fmt.Errorf("restoring the original resolv.conf file return err: %w", err)
 			}
 		}
 		return fmt.Errorf("unable to configure DNS for this peer using file manager without a nameserver group with all domains configured")
 	}
 
 	if !backupFileExist {
-		err = f.backup()
+		err := f.backup()
 		if err != nil {
-			return fmt.Errorf("unable to backup the resolv.conf file")
+			return fmt.Errorf("unable to backup the resolv.conf file: %w", err)
 		}
 	}
 
 	nbSearchDomains := searchDomains(config)
-	nbNameserverIP := config.ServerIP
+	f.nbNameserverIP = config.ServerIP
 
 	resolvConf, err := parseBackupResolvConf()
 	if err != nil {
-		log.Error(err)
+		log.Errorf("could not read original search domains from %s: %s", fileDefaultResolvConfBackupLocation, err)
 	}
 
 	f.repair.stopWatchFileChanges()
 
-	err = f.updateConfig(nbSearchDomains, nbNameserverIP, resolvConf)
+	err = f.updateConfig(nbSearchDomains, f.nbNameserverIP, resolvConf)
 	if err != nil {
 		return err
 	}
-	f.repair.watchFileChanges(nbSearchDomains, nbNameserverIP)
+	f.repair.watchFileChanges(nbSearchDomains, f.nbNameserverIP)
 	return nil
 }
 
@@ -84,10 +88,11 @@ func (f *fileConfigurator) updateConfig(nbSearchDomains []string, nbNameserverIP
 	searchDomainList := mergeSearchDomains(nbSearchDomains, cfg.searchDomains)
 	nameServers := generateNsList(nbNameserverIP, cfg)
 
+	options := prepareOptionsWithTimeout(cfg.others, int(dnsFailoverTimeout.Seconds()), dnsFailoverAttempts)
 	buf := prepareResolvConfContent(
 		searchDomainList,
 		nameServers,
-		cfg.others)
+		options)
 
 	log.Debugf("creating managed file %s", defaultResolvConfPath)
 	err := os.WriteFile(defaultResolvConfPath, buf.Bytes(), f.originalPerms)
@@ -96,10 +101,16 @@ func (f *fileConfigurator) updateConfig(nbSearchDomains []string, nbNameserverIP
 		if restoreErr != nil {
 			log.Errorf("attempt to restore default file failed with error: %s", err)
 		}
-		return fmt.Errorf("got an creating resolver file %s. Error: %s", defaultResolvConfPath, err)
+		return fmt.Errorf("creating resolver file %s. Error: %w", defaultResolvConfPath, err)
+	}
+
+	log.Infof("created a NetBird managed %s file with the DNS settings. Added %d search domains. Search list: %s", defaultResolvConfPath, len(searchDomainList), searchDomainList)
+
+	// create another backup for unclean shutdown detection right after overwriting the original resolv.conf
+	if err := createUncleanShutdownIndicator(fileDefaultResolvConfBackupLocation, fileManager, nbNameserverIP); err != nil {
+		log.Errorf("failed to create unclean shutdown resolv.conf backup: %s", err)
 	}
 
-	log.Infof("created a NetBird managed %s file with your DNS settings. Added %d search domains. Search list: %s", defaultResolvConfPath, len(searchDomainList), searchDomainList)
 	return nil
 }
 
@@ -111,27 +122,83 @@ func (f *fileConfigurator) restoreHostDNS() error {
 func (f *fileConfigurator) backup() error {
 	stats, err := os.Stat(defaultResolvConfPath)
 	if err != nil {
-		return fmt.Errorf("got an error while checking stats for %s file. Error: %s", defaultResolvConfPath, err)
+		return fmt.Errorf("checking stats for %s file. Error: %w", defaultResolvConfPath, err)
 	}
 
 	f.originalPerms = stats.Mode()
 
 	err = copyFile(defaultResolvConfPath, fileDefaultResolvConfBackupLocation)
 	if err != nil {
-		return fmt.Errorf("got error while backing up the %s file. Error: %s", defaultResolvConfPath, err)
+		return fmt.Errorf("backing up %s: %w", defaultResolvConfPath, err)
 	}
 	return nil
 }
 
 func (f *fileConfigurator) restore() error {
-	err := copyFile(fileDefaultResolvConfBackupLocation, defaultResolvConfPath)
+	err := removeFirstNbNameserver(fileDefaultResolvConfBackupLocation, f.nbNameserverIP)
 	if err != nil {
-		return fmt.Errorf("got error while restoring the %s file from %s. Error: %s", defaultResolvConfPath, fileDefaultResolvConfBackupLocation, err)
+		log.Errorf("Failed to remove netbird nameserver from %s on backup restore: %s", fileDefaultResolvConfBackupLocation, err)
+	}
+
+	err = copyFile(fileDefaultResolvConfBackupLocation, defaultResolvConfPath)
+	if err != nil {
+		return fmt.Errorf("restoring %s from %s: %w", defaultResolvConfPath, fileDefaultResolvConfBackupLocation, err)
+	}
+
+	if err := removeUncleanShutdownIndicator(); err != nil {
+		log.Errorf("failed to remove unclean shutdown resolv.conf backup: %s", err)
 	}
 
 	return os.RemoveAll(fileDefaultResolvConfBackupLocation)
 }
 
+func (f *fileConfigurator) restoreUncleanShutdownDNS(storedDNSAddress *netip.Addr) error {
+	resolvConf, err := parseDefaultResolvConf()
+	if err != nil {
+		return fmt.Errorf("parse current resolv.conf: %w", err)
+	}
+
+	// no current nameservers set -> restore
+	if len(resolvConf.nameServers) == 0 {
+		return restoreResolvConfFile()
+	}
+
+	currentDNSAddress, err := netip.ParseAddr(resolvConf.nameServers[0])
+	// not a valid first nameserver -> restore
+	if err != nil {
+		log.Errorf("restoring unclean shutdown: parse dns address %s failed: %s", resolvConf.nameServers[0], err)
+		return restoreResolvConfFile()
+	}
+
+	// current address is still netbird's non-available dns address -> restore
+	// comparing parsed addresses only, to remove ambiguity
+	if currentDNSAddress.String() == storedDNSAddress.String() {
+		return restoreResolvConfFile()
+	}
+
+	log.Info("restoring unclean shutdown: first current nameserver differs from saved nameserver pre-netbird: not restoring")
+	return nil
+}
+
+func (f *fileConfigurator) isBackupFileExist() bool {
+	_, err := os.Stat(fileDefaultResolvConfBackupLocation)
+	return err == nil
+}
+
+func restoreResolvConfFile() error {
+	log.Debugf("restoring unclean shutdown: restoring %s from %s", defaultResolvConfPath, fileUncleanShutdownResolvConfLocation)
+
+	if err := copyFile(fileUncleanShutdownResolvConfLocation, defaultResolvConfPath); err != nil {
+		return fmt.Errorf("restoring %s from %s: %w", defaultResolvConfPath, fileUncleanShutdownResolvConfLocation, err)
+	}
+
+	if err := removeUncleanShutdownIndicator(); err != nil {
+		log.Errorf("failed to remove unclean shutdown resolv.conf file: %s", err)
+	}
+
+	return nil
+}
+
 // generateNsList generates a list of nameservers from the config and adds the primary nameserver to the beginning of the list
 func generateNsList(nbNameserverIP string, cfg *resolvConf) []string {
 	ns := make([]string, 1, len(cfg.nameServers)+1)
@@ -231,17 +298,17 @@ func validateAndFillSearchDomains(initialLineChars int, s *[]string, vs []string
 func copyFile(src, dest string) error {
 	stats, err := os.Stat(src)
 	if err != nil {
-		return fmt.Errorf("got an error while checking stats for %s file when copying it. Error: %s", src, err)
+		return fmt.Errorf("checking stats for %s file when copying it. Error: %s", src, err)
 	}
 
 	bytesRead, err := os.ReadFile(src)
 	if err != nil {
-		return fmt.Errorf("got an error while reading the file %s file for copy. Error: %s", src, err)
+		return fmt.Errorf("reading the file %s file for copy. Error: %s", src, err)
 	}
 
 	err = os.WriteFile(dest, bytesRead, stats.Mode())
 	if err != nil {
-		return fmt.Errorf("got an writing the destination file %s for copy. Error: %s", dest, err)
+		return fmt.Errorf("writing the destination file %s for copy. Error: %s", dest, err)
 	}
 	return nil
 }

client/internal/dns/file_parser_linux.go

@@ -5,6 +5,7 @@ package dns
 import (
 	"fmt"
 	"os"
+	"regexp"
 	"strings"
 
 	log "github.com/sirupsen/logrus"
@@ -14,6 +15,9 @@ const (
 	defaultResolvConfPath = "/etc/resolv.conf"
 )
 
+var timeoutRegex = regexp.MustCompile(`timeout:\d+`)
+var attemptsRegex = regexp.MustCompile(`attempts:\d+`)
+
 type resolvConf struct {
 	nameServers   []string
 	searchDomains []string
@@ -33,9 +37,15 @@ func parseBackupResolvConf() (*resolvConf, error) {
 }
 
 func parseResolvConfFile(resolvConfFile string) (*resolvConf, error) {
+	rconf := &resolvConf{
+		searchDomains: make([]string, 0),
+		nameServers:   make([]string, 0),
+		others:        make([]string, 0),
+	}
+
 	file, err := os.Open(resolvConfFile)
 	if err != nil {
-		return nil, fmt.Errorf("failed to open %s file: %w", resolvConfFile, err)
+		return rconf, fmt.Errorf("failed to open %s file: %w", resolvConfFile, err)
 	}
 	defer func() {
 		if err := file.Close(); err != nil {
@@ -45,17 +55,11 @@ func parseResolvConfFile(resolvConfFile string) (*resolvConf, error) {
 
 	cur, err := os.ReadFile(resolvConfFile)
 	if err != nil {
-		return nil, fmt.Errorf("failed to read %s file: %w", resolvConfFile, err)
+		return rconf, fmt.Errorf("failed to read %s file: %w", resolvConfFile, err)
 	}
 
 	if len(cur) == 0 {
-		return nil, fmt.Errorf("file is empty")
-	}
-
-	rconf := &resolvConf{
-		searchDomains: make([]string, 0),
-		nameServers:   make([]string, 0),
-		others:        make([]string, 0),
+		return rconf, fmt.Errorf("file is empty")
 	}
 
 	for _, line := range strings.Split(string(cur), "\n") {
@@ -103,3 +107,62 @@ func parseResolvConfFile(resolvConfFile string) (*resolvConf, error) {
 	}
 	return rconf, nil
 }
+
+// prepareOptionsWithTimeout appends timeout to existing options if it doesn't exist,
+// otherwise it adds a new option with timeout and attempts.
+func prepareOptionsWithTimeout(input []string, timeout int, attempts int) []string {
+	configs := make([]string, len(input))
+	copy(configs, input)
+
+	for i, config := range configs {
+		if strings.HasPrefix(config, "options") {
+			config = strings.ReplaceAll(config, "rotate", "")
+			config = strings.Join(strings.Fields(config), " ")
+
+			if strings.Contains(config, "timeout:") {
+				config = timeoutRegex.ReplaceAllString(config, fmt.Sprintf("timeout:%d", timeout))
+			} else {
+				config = strings.Replace(config, "options ", fmt.Sprintf("options timeout:%d ", timeout), 1)
+			}
+
+			if strings.Contains(config, "attempts:") {
+				config = attemptsRegex.ReplaceAllString(config, fmt.Sprintf("attempts:%d", attempts))
+			} else {
+				config = strings.Replace(config, "options ", fmt.Sprintf("options attempts:%d ", attempts), 1)
+			}
+
+			configs[i] = config
+			return configs
+		}
+	}
+
+	return append(configs, fmt.Sprintf("options timeout:%d attempts:%d", timeout, attempts))
+}
+
+// removeFirstNbNameserver removes the given nameserver from the given file if it is in the first position
+// and writes the file back to the original location
+func removeFirstNbNameserver(filename, nameserverIP string) error {
+	resolvConf, err := parseResolvConfFile(filename)
+	if err != nil {
+		return fmt.Errorf("parse backup resolv.conf: %w", err)
+	}
+	content, err := os.ReadFile(filename)
+	if err != nil {
+		return fmt.Errorf("read %s: %w", filename, err)
+	}
+
+	if len(resolvConf.nameServers) > 1 && resolvConf.nameServers[0] == nameserverIP {
+		newContent := strings.Replace(string(content), fmt.Sprintf("nameserver %s\n", nameserverIP), "", 1)
+
+		stat, err := os.Stat(filename)
+		if err != nil {
+			return fmt.Errorf("stat %s: %w", filename, err)
+		}
+		if err := os.WriteFile(filename, []byte(newContent), stat.Mode()); err != nil {
+			return fmt.Errorf("write %s: %w", filename, err)
+		}
+
+	}
+
+	return nil
+}

client/internal/dns/file_parser_linux_test.go

@@ -3,9 +3,11 @@
 package dns
 
 import (
-	"fmt"
 	"os"
+	"path/filepath"
 	"testing"
+
+	"github.com/stretchr/testify/assert"
 )
 
 func Test_parseResolvConf(t *testing.T) {
@@ -16,11 +18,11 @@ func Test_parseResolvConf(t *testing.T) {
 		expectedOther  []string
 	}{
 		{
-			input: `domain chello.hu
-search chello.hu
+			input: `domain example.org
+search example.org
 nameserver 192.168.0.1
 `,
-			expectedSearch: []string{"chello.hu"},
+			expectedSearch: []string{"example.org"},
 			expectedNS:     []string{"192.168.0.1"},
 			expectedOther:  []string{},
 		},
@@ -75,40 +77,13 @@ options debug
 			expectedNS:     []string{"192.168.2.1", "100.81.99.197"},
 			expectedOther:  []string{"options debug"},
 		},
-		{
-			input: `# This is /run/systemd/resolve/resolv.conf managed by man:systemd-resolved(8).
-# Do not edit.
-#
-# This file might be symlinked as /etc/resolv.conf. If you're looking at
-# /etc/resolv.conf and seeing this text, you have followed the symlink.
-#
-# This is a dynamic resolv.conf file for connecting local clients directly to
-# all known uplink DNS servers. This file lists all configured search domains.
-#
-# Third party programs should typically not access this file directly, but only
-# through the symlink at /etc/resolv.conf. To manage man:resolv.conf(5) in a
-# different way, replace this symlink by a static file or a different symlink.
-#
-# See man:systemd-resolved.service(8) for details about the supported modes of
-# operation for /etc/resolv.conf.
-
-nameserver 192.168.2.1
-nameserver 100.81.99.197
-search netbird.cloud
-options debug
-options edns0 trust-ad
-`,
-			expectedSearch: []string{"netbird.cloud"},
-			expectedNS:     []string{"192.168.2.1", "100.81.99.197"},
-			expectedOther:  []string{"options debug", "options edns0 trust-ad"},
-		},
 	}
 
 	for _, testCase := range testCases {
 		testCase := testCase
 		t.Run("test", func(t *testing.T) {
 			t.Parallel()
-			tmpResolvConf := fmt.Sprintf("%s/%s", t.TempDir(), "resolv.conf")
+			tmpResolvConf := filepath.Join(t.TempDir(), "resolv.conf")
 			err := os.WriteFile(tmpResolvConf, []byte(testCase.input), 0644)
 			if err != nil {
 				t.Fatal(err)
@@ -147,3 +122,183 @@ func compareLists(search []string, search2 []string) bool {
 	}
 	return true
 }
+
+func Test_emptyFile(t *testing.T) {
+	cfg, err := parseResolvConfFile("/tmp/nothing")
+	if err == nil {
+		t.Errorf("expected error, got nil")
+	}
+	if len(cfg.others) != 0 || len(cfg.searchDomains) != 0 || len(cfg.nameServers) != 0 {
+		t.Errorf("expected empty config, got %v", cfg)
+	}
+}
+
+func Test_symlink(t *testing.T) {
+	input := `# This is /run/systemd/resolve/resolv.conf managed by man:systemd-resolved(8).
+# Do not edit.
+#
+# This file might be symlinked as /etc/resolv.conf. If you're looking at
+# /etc/resolv.conf and seeing this text, you have followed the symlink.
+#
+# This is a dynamic resolv.conf file for connecting local clients directly to
+# all known uplink DNS servers. This file lists all configured search domains.
+#
+# Third party programs should typically not access this file directly, but only
+# through the symlink at /etc/resolv.conf. To manage man:resolv.conf(5) in a
+# different way, replace this symlink by a static file or a different symlink.
+#
+# See man:systemd-resolved.service(8) for details about the supported modes of
+# operation for /etc/resolv.conf.
+
+nameserver 192.168.0.1
+`
+
+	tmpResolvConf := filepath.Join(t.TempDir(), "resolv.conf")
+	err := os.WriteFile(tmpResolvConf, []byte(input), 0644)
+	if err != nil {
+		t.Fatal(err)
+	}
+
+	tmpLink := filepath.Join(t.TempDir(), "symlink")
+	err = os.Symlink(tmpResolvConf, tmpLink)
+	if err != nil {
+		t.Fatal(err)
+	}
+
+	cfg, err := parseResolvConfFile(tmpLink)
+	if err != nil {
+		t.Fatal(err)
+	}
+
+	if len(cfg.nameServers) != 1 {
+		t.Errorf("unexpected resolv.conf content: %v", cfg)
+	}
+}
+
+func TestPrepareOptionsWithTimeout(t *testing.T) {
+	tests := []struct {
+		name     string
+		others   []string
+		timeout  int
+		attempts int
+		expected []string
+	}{
+		{
+			name:     "Append new options with timeout and attempts",
+			others:   []string{"some config"},
+			timeout:  2,
+			attempts: 2,
+			expected: []string{"some config", "options timeout:2 attempts:2"},
+		},
+		{
+			name:     "Modify existing options to exclude rotate and include timeout and attempts",
+			others:   []string{"some config", "options rotate someother"},
+			timeout:  3,
+			attempts: 2,
+			expected: []string{"some config", "options attempts:2 timeout:3 someother"},
+		},
+		{
+			name:     "Existing options with timeout and attempts are updated",
+			others:   []string{"some config", "options timeout:4 attempts:3"},
+			timeout:  5,
+			attempts: 4,
+			expected: []string{"some config", "options timeout:5 attempts:4"},
+		},
+		{
+			name:     "Modify existing options, add missing attempts before timeout",
+			others:   []string{"some config", "options timeout:4"},
+			timeout:  4,
+			attempts: 3,
+			expected: []string{"some config", "options attempts:3 timeout:4"},
+		},
+	}
+
+	for _, tc := range tests {
+		t.Run(tc.name, func(t *testing.T) {
+			result := prepareOptionsWithTimeout(tc.others, tc.timeout, tc.attempts)
+			assert.Equal(t, tc.expected, result)
+		})
+	}
+}
+
+func TestRemoveFirstNbNameserver(t *testing.T) {
+	testCases := []struct {
+		name       string
+		content    string
+		ipToRemove string
+		expected   string
+	}{
+		{
+			name: "Unrelated nameservers with comments and options",
+			content: `# This is a comment
+options rotate
+nameserver 1.1.1.1
+# Another comment
+nameserver 8.8.4.4
+search example.com`,
+			ipToRemove: "9.9.9.9",
+			expected: `# This is a comment
+options rotate
+nameserver 1.1.1.1
+# Another comment
+nameserver 8.8.4.4
+search example.com`,
+		},
+		{
+			name: "First nameserver matches",
+			content: `search example.com
+nameserver 9.9.9.9
+# oof, a comment
+nameserver 8.8.4.4
+options attempts:5`,
+			ipToRemove: "9.9.9.9",
+			expected: `search example.com
+# oof, a comment
+nameserver 8.8.4.4
+options attempts:5`,
+		},
+		{
+			name: "Target IP not the first nameserver",
+			// nolint:dupword
+			content: `# Comment about the first nameserver
+nameserver 8.8.4.4
+# Comment before our target
+nameserver 9.9.9.9
+options timeout:2`,
+			ipToRemove: "9.9.9.9",
+			// nolint:dupword
+			expected: `# Comment about the first nameserver
+nameserver 8.8.4.4
+# Comment before our target
+nameserver 9.9.9.9
+options timeout:2`,
+		},
+		{
+			name: "Only nameserver matches",
+			content: `options debug
+nameserver 9.9.9.9
+search localdomain`,
+			ipToRemove: "9.9.9.9",
+			expected: `options debug
+nameserver 9.9.9.9
+search localdomain`,
+		},
+	}
+
+	for _, tc := range testCases {
+		t.Run(tc.name, func(t *testing.T) {
+			tempDir := t.TempDir()
+			tempFile := filepath.Join(tempDir, "resolv.conf")
+			err := os.WriteFile(tempFile, []byte(tc.content), 0644)
+			assert.NoError(t, err)
+
+			err = removeFirstNbNameserver(tempFile, tc.ipToRemove)
+			assert.NoError(t, err)
+
+			content, err := os.ReadFile(tempFile)
+			assert.NoError(t, err)
+
+			assert.Equal(t, tc.expected, string(content), "The resulting content should match the expected output.")
+		})
+	}
+}

client/internal/dns/file_repair_linux.go

@@ -3,8 +3,8 @@
 package dns
 
 import (
-	"fmt"
 	"path"
+	"path/filepath"
 	"sync"
 
 	"github.com/fsnotify/fsnotify"
@@ -32,9 +32,10 @@ type repair struct {
 }
 
 func newRepair(operationFile string, updateFn repairConfFn) *repair {
+	targetFile := targetFile(operationFile)
 	return &repair{
-		operationFile: operationFile,
-		watchDir:      path.Dir(operationFile),
+		operationFile: targetFile,
+		watchDir:      path.Dir(targetFile),
 		updateFn:      updateFn,
 	}
 }
@@ -44,7 +45,7 @@ func (f *repair) watchFileChanges(nbSearchDomains []string, nbNameserverIP strin
 		return
 	}
 
-	log.Infof("start to watch resolv.conf")
+	log.Infof("start to watch resolv.conf: %s", f.operationFile)
 	inotify, err := fsnotify.NewWatcher()
 	if err != nil {
 		log.Errorf("failed to start inotify watcher for resolv.conf: %s", err)
@@ -60,7 +61,7 @@ func (f *repair) watchFileChanges(nbSearchDomains []string, nbNameserverIP strin
 				continue
 			}
 
-			log.Tracef("resolv.conf changed, check if it is broken")
+			log.Tracef("%s changed, check if it is broken", f.operationFile)
 
 			rConf, err := parseResolvConfFile(f.operationFile)
 			if err != nil {
@@ -87,7 +88,7 @@ func (f *repair) watchFileChanges(nbSearchDomains []string, nbNameserverIP strin
 
 			err = f.inotify.Add(f.watchDir)
 			if err != nil {
-				log.Errorf("failed to readd inotify watch for resolv.conf: %s", err)
+				log.Errorf("failed to re-add inotify watch for resolv.conf: %s", err)
 				return
 			}
 		}
@@ -124,8 +125,7 @@ func (f *repair) isEventRelevant(event fsnotify.Event) bool {
 		return false
 	}
 
-	operationFileSymlink := fmt.Sprintf("%s~", f.operationFile)
-	if event.Name == f.operationFile || event.Name == operationFileSymlink {
+	if event.Name == f.operationFile {
 		return true
 	}
 	return false
@@ -149,3 +149,11 @@ func isNbParamsMissing(nbSearchDomains []string, nbNameserverIP string, rConf *r
 
 	return false
 }
+
+func targetFile(filename string) string {
+	target, err := filepath.EvalSymlinks(filename)
+	if err != nil {
+		log.Errorf("evarl err: %s", err)
+	}
+	return target
+}

client/internal/dns/file_repair_linux_test.go

@@ -5,6 +5,7 @@ package dns
 import (
 	"context"
 	"os"
+	"path/filepath"
 	"testing"
 	"time"
 
@@ -126,5 +127,49 @@ nameserver 8.8.8.8`,
 			}
 		})
 	}
-
+}
+
+func Test_newRepairSymlink(t *testing.T) {
+	resolvConfContent := `
+nameserver 10.0.0.1
+nameserver 8.8.8.8
+searchdomain netbird.cloud something`
+
+	modifyContent := `nameserver 8.8.8.8`
+
+	tmpResolvConf := filepath.Join(t.TempDir(), "resolv.conf")
+	err := os.WriteFile(tmpResolvConf, []byte(resolvConfContent), 0644)
+	if err != nil {
+		t.Fatal(err)
+	}
+
+	tmpLink := filepath.Join(t.TempDir(), "symlink")
+	err = os.Symlink(tmpResolvConf, tmpLink)
+	if err != nil {
+		t.Fatal(err)
+	}
+
+	var changed bool
+	ctx, cancel := context.WithTimeout(context.Background(), time.Second)
+	updateFn := func([]string, string, *resolvConf) error {
+		changed = true
+		cancel()
+		return nil
+	}
+
+	r := newRepair(tmpLink, updateFn)
+	r.watchFileChanges([]string{"netbird.cloud"}, "10.0.0.1")
+
+	err = os.WriteFile(tmpLink, []byte(modifyContent), 0755)
+	if err != nil {
+		t.Fatalf("failed to write out resolv.conf: %s", err)
+	}
+
+	<-ctx.Done()
+
+	r.stopWatchFileChanges()
+
+	if changed != true {
+		t.Errorf("unexpected result: want: %v, got: %v", true, false)
+	}
 }

client/internal/dns/host.go

@@ -2,6 +2,7 @@ package dns
 
 import (
 	"fmt"
+	"net/netip"
 	"strings"
 
 	nbdns "github.com/netbirdio/netbird/dns"
@@ -11,6 +12,7 @@ type hostManager interface {
 	applyDNSConfig(config HostDNSConfig) error
 	restoreHostDNS() error
 	supportCustomPort() bool
+	restoreUncleanShutdownDNS(storedDNSAddress *netip.Addr) error
 }
 
 type HostDNSConfig struct {
@@ -27,9 +29,10 @@ type DomainConfig struct {
 }
 
 type mockHostConfigurator struct {
-	applyDNSConfigFunc    func(config HostDNSConfig) error
-	restoreHostDNSFunc    func() error
-	supportCustomPortFunc func() bool
+	applyDNSConfigFunc            func(config HostDNSConfig) error
+	restoreHostDNSFunc            func() error
+	supportCustomPortFunc         func() bool
+	restoreUncleanShutdownDNSFunc func(*netip.Addr) error
 }
 
 func (m *mockHostConfigurator) applyDNSConfig(config HostDNSConfig) error {
@@ -53,11 +56,19 @@ func (m *mockHostConfigurator) supportCustomPort() bool {
 	return false
 }
 
+func (m *mockHostConfigurator) restoreUncleanShutdownDNS(storedDNSAddress *netip.Addr) error {
+	if m.restoreUncleanShutdownDNSFunc != nil {
+		return m.restoreUncleanShutdownDNSFunc(storedDNSAddress)
+	}
+	return fmt.Errorf("method restoreUncleanShutdownDNS is not implemented")
+}
+
 func newNoopHostMocker() hostManager {
 	return &mockHostConfigurator{
-		applyDNSConfigFunc:    func(config HostDNSConfig) error { return nil },
-		restoreHostDNSFunc:    func() error { return nil },
-		supportCustomPortFunc: func() bool { return true },
+		applyDNSConfigFunc:            func(config HostDNSConfig) error { return nil },
+		restoreHostDNSFunc:            func() error { return nil },
+		supportCustomPortFunc:         func() bool { return true },
+		restoreUncleanShutdownDNSFunc: func(*netip.Addr) error { return nil },
 	}
 }
 

client/internal/dns/host_android.go

@@ -1,9 +1,11 @@
 package dns
 
+import "net/netip"
+
 type androidHostManager struct {
 }
 
-func newHostManager(wgInterface WGIface) (hostManager, error) {
+func newHostManager() (hostManager, error) {
 	return &androidHostManager{}, nil
 }
 
@@ -18,3 +20,7 @@ func (a androidHostManager) restoreHostDNS() error {
 func (a androidHostManager) supportCustomPort() bool {
 	return false
 }
+
+func (a androidHostManager) restoreUncleanShutdownDNS(*netip.Addr) error {
+	return nil
+}

client/internal/dns/host_darwin.go

@@ -6,6 +6,8 @@ import (
 	"bufio"
 	"bytes"
 	"fmt"
+	"io"
+	"net/netip"
 	"os/exec"
 	"strconv"
 	"strings"
@@ -34,7 +36,7 @@ type systemConfigurator struct {
 	createdKeys      map[string]struct{}
 }
 
-func newHostManager(_ WGIface) (hostManager, error) {
+func newHostManager() (hostManager, error) {
 	return &systemConfigurator{
 		createdKeys: make(map[string]struct{}),
 	}, nil
@@ -50,17 +52,22 @@ func (s *systemConfigurator) applyDNSConfig(config HostDNSConfig) error {
 	if config.RouteAll {
 		err = s.addDNSSetupForAll(config.ServerIP, config.ServerPort)
 		if err != nil {
-			return err
+			return fmt.Errorf("add dns setup for all: %w", err)
 		}
 	} else if s.primaryServiceID != "" {
 		err = s.removeKeyFromSystemConfig(getKeyWithInput(primaryServiceSetupKeyFormat, s.primaryServiceID))
 		if err != nil {
-			return err
+			return fmt.Errorf("remote key from system config: %w", err)
 		}
 		s.primaryServiceID = ""
 		log.Infof("removed %s:%d as main DNS resolver for this peer", config.ServerIP, config.ServerPort)
 	}
 
+	// create a file for unclean shutdown detection
+	if err := createUncleanShutdownIndicator(); err != nil {
+		log.Errorf("failed to create unclean shutdown file: %s", err)
+	}
+
 	var (
 		searchDomains []string
 		matchDomains  []string
@@ -85,7 +92,7 @@ func (s *systemConfigurator) applyDNSConfig(config HostDNSConfig) error {
 		err = s.removeKeyFromSystemConfig(matchKey)
 	}
 	if err != nil {
-		return err
+		return fmt.Errorf("add match domains: %w", err)
 	}
 
 	searchKey := getKeyWithInput(netbirdDNSStateKeyFormat, searchSuffix)
@@ -96,7 +103,7 @@ func (s *systemConfigurator) applyDNSConfig(config HostDNSConfig) error {
 		err = s.removeKeyFromSystemConfig(searchKey)
 	}
 	if err != nil {
-		return err
+		return fmt.Errorf("add search domains: %w", err)
 	}
 
 	return nil
@@ -119,7 +126,11 @@ func (s *systemConfigurator) restoreHostDNS() error {
 	_, err := runSystemConfigCommand(wrapCommand(lines))
 	if err != nil {
 		log.Errorf("got an error while cleaning the system configuration: %s", err)
-		return err
+		return fmt.Errorf("clean system: %w", err)
+	}
+
+	if err := removeUncleanShutdownIndicator(); err != nil {
+		log.Errorf("failed to remove unclean shutdown file: %s", err)
 	}
 
 	return nil
@@ -129,7 +140,7 @@ func (s *systemConfigurator) removeKeyFromSystemConfig(key string) error {
 	line := buildRemoveKeyOperation(key)
 	_, err := runSystemConfigCommand(wrapCommand(line))
 	if err != nil {
-		return err
+		return fmt.Errorf("remove key: %w", err)
 	}
 
 	delete(s.createdKeys, key)
@@ -140,7 +151,7 @@ func (s *systemConfigurator) removeKeyFromSystemConfig(key string) error {
 func (s *systemConfigurator) addSearchDomains(key, domains string, ip string, port int) error {
 	err := s.addDNSState(key, domains, ip, port, true)
 	if err != nil {
-		return err
+		return fmt.Errorf("add dns state: %w", err)
 	}
 
 	log.Infof("added %d search domains to the state. Domain list: %s", len(strings.Split(domains, " ")), domains)
@@ -153,7 +164,7 @@ func (s *systemConfigurator) addSearchDomains(key, domains string, ip string, po
 func (s *systemConfigurator) addMatchDomains(key, domains, dnsServer string, port int) error {
 	err := s.addDNSState(key, domains, dnsServer, port, false)
 	if err != nil {
-		return err
+		return fmt.Errorf("add dns state: %w", err)
 	}
 
 	log.Infof("added %d match domains to the state. Domain list: %s", len(strings.Split(domains, " ")), domains)
@@ -178,33 +189,37 @@ func (s *systemConfigurator) addDNSState(state, domains, dnsServer string, port
 
 	_, err := runSystemConfigCommand(stdinCommands)
 	if err != nil {
-		return fmt.Errorf("got error while applying state for domains %s, error: %s", domains, err)
+		return fmt.Errorf("applying state for domains %s, error: %w", domains, err)
 	}
 	return nil
 }
 
 func (s *systemConfigurator) addDNSSetupForAll(dnsServer string, port int) error {
-	primaryServiceKey, existingNameserver := s.getPrimaryService()
-	if primaryServiceKey == "" {
-		return fmt.Errorf("couldn't find the primary service key")
+	primaryServiceKey, existingNameserver, err := s.getPrimaryService()
+	if err != nil || primaryServiceKey == "" {
+		return fmt.Errorf("couldn't find the primary service key: %w", err)
 	}
-	err := s.addDNSSetup(getKeyWithInput(primaryServiceSetupKeyFormat, primaryServiceKey), dnsServer, port, existingNameserver)
+
+	err = s.addDNSSetup(getKeyWithInput(primaryServiceSetupKeyFormat, primaryServiceKey), dnsServer, port, existingNameserver)
 	if err != nil {
-		return err
+		return fmt.Errorf("add dns setup: %w", err)
 	}
+
 	log.Infof("configured %s:%d as main DNS resolver for this peer", dnsServer, port)
 	s.primaryServiceID = primaryServiceKey
+
 	return nil
 }
 
-func (s *systemConfigurator) getPrimaryService() (string, string) {
+func (s *systemConfigurator) getPrimaryService() (string, string, error) {
 	line := buildCommandLine("show", globalIPv4State, "")
 	stdinCommands := wrapCommand(line)
+
 	b, err := runSystemConfigCommand(stdinCommands)
 	if err != nil {
-		log.Error("got error while sending the command: ", err)
-		return "", ""
+		return "", "", fmt.Errorf("sending the command: %w", err)
 	}
+
 	scanner := bufio.NewScanner(bytes.NewReader(b))
 	primaryService := ""
 	router := ""
@@ -217,7 +232,11 @@ func (s *systemConfigurator) getPrimaryService() (string, string) {
 			router = strings.TrimSpace(strings.Split(text, ":")[1])
 		}
 	}
-	return primaryService, router
+	if err := scanner.Err(); err != nil && err != io.EOF {
+		return primaryService, router, fmt.Errorf("scan: %w", err)
+	}
+
+	return primaryService, router, nil
 }
 
 func (s *systemConfigurator) addDNSSetup(setupKey, dnsServer string, port int, existingDNSServer string) error {
@@ -228,7 +247,14 @@ func (s *systemConfigurator) addDNSSetup(setupKey, dnsServer string, port int, e
 	stdinCommands := wrapCommand(addDomainCommand)
 	_, err := runSystemConfigCommand(stdinCommands)
 	if err != nil {
-		return fmt.Errorf("got error while applying dns setup, error: %s", err)
+		return fmt.Errorf("applying dns setup, error: %w", err)
+	}
+	return nil
+}
+
+func (s *systemConfigurator) restoreUncleanShutdownDNS(*netip.Addr) error {
+	if err := s.restoreHostDNS(); err != nil {
+		return fmt.Errorf("restoring dns via scutil: %w", err)
 	}
 	return nil
 }
@@ -266,7 +292,7 @@ func runSystemConfigCommand(command string) ([]byte, error) {
 	cmd.Stdin = strings.NewReader(command)
 	out, err := cmd.Output()
 	if err != nil {
-		return nil, fmt.Errorf("got error while running system configuration command: \"%s\", error: %s", command, err)
+		return nil, fmt.Errorf("running system configuration command: \"%s\", error: %w", command, err)
 	}
 	return out, nil
 }

client/internal/dns/host_ios.go

@@ -2,6 +2,8 @@ package dns
 
 import (
 	"encoding/json"
+	"fmt"
+	"net/netip"
 
 	log "github.com/sirupsen/logrus"
 )
@@ -20,7 +22,7 @@ func newHostManager(dnsManager IosDnsManager) (hostManager, error) {
 func (a iosHostManager) applyDNSConfig(config HostDNSConfig) error {
 	jsonData, err := json.Marshal(config)
 	if err != nil {
-		return err
+		return fmt.Errorf("marshal: %w", err)
 	}
 	jsonString := string(jsonData)
 	log.Debugf("Applying DNS settings: %s", jsonString)
@@ -35,3 +37,7 @@ func (a iosHostManager) restoreHostDNS() error {
 func (a iosHostManager) supportCustomPort() bool {
 	return false
 }
+
+func (a iosHostManager) restoreUncleanShutdownDNS(*netip.Addr) error {
+	return nil
+}

client/internal/dns/host_linux.go

@@ -4,7 +4,9 @@ package dns
 
 import (
 	"bufio"
+	"errors"
 	"fmt"
+	"io"
 	"os"
 	"strings"
 
@@ -19,8 +21,27 @@ const (
 	resolvConfManager
 )
 
+var ErrUnknownOsManagerType = errors.New("unknown os manager type")
+
 type osManagerType int
 
+func newOsManagerType(osManager string) (osManagerType, error) {
+	switch osManager {
+	case "netbird":
+		return fileManager, nil
+	case "file":
+		return netbirdManager, nil
+	case "networkManager":
+		return networkManager, nil
+	case "systemd":
+		return systemdManager, nil
+	case "resolvconf":
+		return resolvConfManager, nil
+	default:
+		return 0, ErrUnknownOsManagerType
+	}
+}
+
 func (t osManagerType) String() string {
 	switch t {
 	case netbirdManager:
@@ -38,13 +59,17 @@ func (t osManagerType) String() string {
 	}
 }
 
-func newHostManager(wgInterface WGIface) (hostManager, error) {
+func newHostManager(wgInterface string) (hostManager, error) {
 	osManager, err := getOSDNSManagerType()
 	if err != nil {
 		return nil, err
 	}
 
-	log.Debugf("discovered mode is: %s", osManager)
+	log.Infof("System DNS manager discovered: %s", osManager)
+	return newHostManagerFromType(wgInterface, osManager)
+}
+
+func newHostManagerFromType(wgInterface string, osManager osManagerType) (hostManager, error) {
 	switch osManager {
 	case networkManager:
 		return newNetworkManagerDbusConfigurator(wgInterface)
@@ -58,12 +83,15 @@ func newHostManager(wgInterface WGIface) (hostManager, error) {
 }
 
 func getOSDNSManagerType() (osManagerType, error) {
-
 	file, err := os.Open(defaultResolvConfPath)
 	if err != nil {
-		return 0, fmt.Errorf("unable to open %s for checking owner, got error: %s", defaultResolvConfPath, err)
+		return 0, fmt.Errorf("unable to open %s for checking owner, got error: %w", defaultResolvConfPath, err)
 	}
-	defer file.Close()
+	defer func() {
+		if err := file.Close(); err != nil {
+			log.Errorf("close file %s: %s", defaultResolvConfPath, err)
+		}
+	}()
 
 	scanner := bufio.NewScanner(file)
 	for scanner.Scan() {
@@ -101,6 +129,10 @@ func getOSDNSManagerType() (osManagerType, error) {
 			return resolvConfManager, nil
 		}
 	}
+	if err := scanner.Err(); err != nil && err != io.EOF {
+		return 0, fmt.Errorf("scan: %w", err)
+	}
+
 	return fileManager, nil
 }
 

client/internal/dns/host_windows.go

@@ -2,6 +2,8 @@ package dns
 
 import (
 	"fmt"
+	"io"
+	"net/netip"
 	"strings"
 
 	log "github.com/sirupsen/logrus"
@@ -9,7 +11,7 @@ import (
 )
 
 const (
-	dnsPolicyConfigMatchPath            = "SYSTEM\\CurrentControlSet\\Services\\Dnscache\\Parameters\\DnsPolicyConfig\\NetBird-Match"
+	dnsPolicyConfigMatchPath            = `SYSTEM\CurrentControlSet\Services\Dnscache\Parameters\DnsPolicyConfig\NetBird-Match`
 	dnsPolicyConfigVersionKey           = "Version"
 	dnsPolicyConfigVersionValue         = 2
 	dnsPolicyConfigNameKey              = "Name"
@@ -19,7 +21,7 @@ const (
 )
 
 const (
-	interfaceConfigPath          = "SYSTEM\\CurrentControlSet\\Services\\Tcpip\\Parameters\\Interfaces"
+	interfaceConfigPath          = `SYSTEM\CurrentControlSet\Services\Tcpip\Parameters\Interfaces`
 	interfaceConfigNameServerKey = "NameServer"
 	interfaceConfigSearchListKey = "SearchList"
 )
@@ -34,12 +36,16 @@ func newHostManager(wgInterface WGIface) (hostManager, error) {
 	if err != nil {
 		return nil, err
 	}
+	return newHostManagerWithGuid(guid)
+}
+
+func newHostManagerWithGuid(guid string) (hostManager, error) {
 	return &registryConfigurator{
 		guid: guid,
 	}, nil
 }
 
-func (s *registryConfigurator) supportCustomPort() bool {
+func (r *registryConfigurator) supportCustomPort() bool {
 	return false
 }
 
@@ -48,17 +54,22 @@ func (r *registryConfigurator) applyDNSConfig(config HostDNSConfig) error {
 	if config.RouteAll {
 		err = r.addDNSSetupForAll(config.ServerIP)
 		if err != nil {
-			return err
+			return fmt.Errorf("add dns setup: %w", err)
 		}
 	} else if r.routingAll {
 		err = r.deleteInterfaceRegistryKeyProperty(interfaceConfigNameServerKey)
 		if err != nil {
-			return err
+			return fmt.Errorf("delete interface registry key property: %w", err)
 		}
 		r.routingAll = false
 		log.Infof("removed %s as main DNS forwarder for this peer", config.ServerIP)
 	}
 
+	// create a file for unclean shutdown detection
+	if err := createUncleanShutdownIndicator(r.guid); err != nil {
+		log.Errorf("failed to create unclean shutdown file: %s", err)
+	}
+
 	var (
 		searchDomains []string
 		matchDomains  []string
@@ -80,12 +91,12 @@ func (r *registryConfigurator) applyDNSConfig(config HostDNSConfig) error {
 		err = removeRegistryKeyFromDNSPolicyConfig(dnsPolicyConfigMatchPath)
 	}
 	if err != nil {
-		return err
+		return fmt.Errorf("add dns match policy: %w", err)
 	}
 
 	err = r.updateSearchDomains(searchDomains)
 	if err != nil {
-		return err
+		return fmt.Errorf("update search domains: %w", err)
 	}
 
 	return nil
@@ -94,7 +105,7 @@ func (r *registryConfigurator) applyDNSConfig(config HostDNSConfig) error {
 func (r *registryConfigurator) addDNSSetupForAll(ip string) error {
 	err := r.setInterfaceRegistryKeyStringValue(interfaceConfigNameServerKey, ip)
 	if err != nil {
-		return fmt.Errorf("adding dns setup for all failed with error: %s", err)
+		return fmt.Errorf("adding dns setup for all failed with error: %w", err)
 	}
 	r.routingAll = true
 	log.Infof("configured %s:53 as main DNS forwarder for this peer", ip)
@@ -106,33 +117,33 @@ func (r *registryConfigurator) addDNSMatchPolicy(domains []string, ip string) er
 	if err == nil {
 		err = registry.DeleteKey(registry.LOCAL_MACHINE, dnsPolicyConfigMatchPath)
 		if err != nil {
-			return fmt.Errorf("unable to remove existing key from registry, key: HKEY_LOCAL_MACHINE\\%s, error: %s", dnsPolicyConfigMatchPath, err)
+			return fmt.Errorf("unable to remove existing key from registry, key: HKEY_LOCAL_MACHINE\\%s, error: %w", dnsPolicyConfigMatchPath, err)
 		}
 	}
 
 	regKey, _, err := registry.CreateKey(registry.LOCAL_MACHINE, dnsPolicyConfigMatchPath, registry.SET_VALUE)
 	if err != nil {
-		return fmt.Errorf("unable to create registry key, key: HKEY_LOCAL_MACHINE\\%s, error: %s", dnsPolicyConfigMatchPath, err)
+		return fmt.Errorf("unable to create registry key, key: HKEY_LOCAL_MACHINE\\%s, error: %w", dnsPolicyConfigMatchPath, err)
 	}
 
 	err = regKey.SetDWordValue(dnsPolicyConfigVersionKey, dnsPolicyConfigVersionValue)
 	if err != nil {
-		return fmt.Errorf("unable to set registry value for %s, error: %s", dnsPolicyConfigVersionKey, err)
+		return fmt.Errorf("unable to set registry value for %s, error: %w", dnsPolicyConfigVersionKey, err)
 	}
 
 	err = regKey.SetStringsValue(dnsPolicyConfigNameKey, domains)
 	if err != nil {
-		return fmt.Errorf("unable to set registry value for %s, error: %s", dnsPolicyConfigNameKey, err)
+		return fmt.Errorf("unable to set registry value for %s, error: %w", dnsPolicyConfigNameKey, err)
 	}
 
 	err = regKey.SetStringValue(dnsPolicyConfigGenericDNSServersKey, ip)
 	if err != nil {
-		return fmt.Errorf("unable to set registry value for %s, error: %s", dnsPolicyConfigGenericDNSServersKey, err)
+		return fmt.Errorf("unable to set registry value for %s, error: %w", dnsPolicyConfigGenericDNSServersKey, err)
 	}
 
 	err = regKey.SetDWordValue(dnsPolicyConfigConfigOptionsKey, dnsPolicyConfigConfigOptionsValue)
 	if err != nil {
-		return fmt.Errorf("unable to set registry value for %s, error: %s", dnsPolicyConfigConfigOptionsKey, err)
+		return fmt.Errorf("unable to set registry value for %s, error: %w", dnsPolicyConfigConfigOptionsKey, err)
 	}
 
 	log.Infof("added %d match domains to the state. Domain list: %s", len(domains), domains)
@@ -141,18 +152,25 @@ func (r *registryConfigurator) addDNSMatchPolicy(domains []string, ip string) er
 }
 
 func (r *registryConfigurator) restoreHostDNS() error {
-	err := removeRegistryKeyFromDNSPolicyConfig(dnsPolicyConfigMatchPath)
-	if err != nil {
-		log.Error(err)
+	if err := removeRegistryKeyFromDNSPolicyConfig(dnsPolicyConfigMatchPath); err != nil {
+		log.Errorf("remove registry key from dns policy config: %s", err)
 	}
 
-	return r.deleteInterfaceRegistryKeyProperty(interfaceConfigSearchListKey)
+	if err := r.deleteInterfaceRegistryKeyProperty(interfaceConfigSearchListKey); err != nil {
+		return fmt.Errorf("remove interface registry key: %w", err)
+	}
+
+	if err := removeUncleanShutdownIndicator(); err != nil {
+		log.Errorf("failed to remove unclean shutdown file: %s", err)
+	}
+
+	return nil
 }
 
 func (r *registryConfigurator) updateSearchDomains(domains []string) error {
 	err := r.setInterfaceRegistryKeyStringValue(interfaceConfigSearchListKey, strings.Join(domains, ","))
 	if err != nil {
-		return fmt.Errorf("adding search domain failed with error: %s", err)
+		return fmt.Errorf("adding search domain failed with error: %w", err)
 	}
 
 	log.Infof("updated the search domains in the registry with %d domains. Domain list: %s", len(domains), domains)
@@ -163,13 +181,13 @@ func (r *registryConfigurator) updateSearchDomains(domains []string) error {
 func (r *registryConfigurator) setInterfaceRegistryKeyStringValue(key, value string) error {
 	regKey, err := r.getInterfaceRegistryKey()
 	if err != nil {
-		return err
+		return fmt.Errorf("get interface registry key: %w", err)
 	}
-	defer regKey.Close()
+	defer closer(regKey)
 
 	err = regKey.SetStringValue(key, value)
 	if err != nil {
-		return fmt.Errorf("applying key %s with value \"%s\" for interface failed with error: %s", key, value, err)
+		return fmt.Errorf("applying key %s with value \"%s\" for interface failed with error: %w", key, value, err)
 	}
 
 	return nil
@@ -178,13 +196,13 @@ func (r *registryConfigurator) setInterfaceRegistryKeyStringValue(key, value str
 func (r *registryConfigurator) deleteInterfaceRegistryKeyProperty(propertyKey string) error {
 	regKey, err := r.getInterfaceRegistryKey()
 	if err != nil {
-		return err
+		return fmt.Errorf("get interface registry key: %w", err)
 	}
-	defer regKey.Close()
+	defer closer(regKey)
 
 	err = regKey.DeleteValue(propertyKey)
 	if err != nil {
-		return fmt.Errorf("deleting registry key %s for interface failed with error: %s", propertyKey, err)
+		return fmt.Errorf("deleting registry key %s for interface failed with error: %w", propertyKey, err)
 	}
 
 	return nil
@@ -197,20 +215,33 @@ func (r *registryConfigurator) getInterfaceRegistryKey() (registry.Key, error) {
 
 	regKey, err := registry.OpenKey(registry.LOCAL_MACHINE, regKeyPath, registry.SET_VALUE)
 	if err != nil {
-		return regKey, fmt.Errorf("unable to open the interface registry key, key: HKEY_LOCAL_MACHINE\\%s, error: %s", regKeyPath, err)
+		return regKey, fmt.Errorf("unable to open the interface registry key, key: HKEY_LOCAL_MACHINE\\%s, error: %w", regKeyPath, err)
 	}
 
 	return regKey, nil
 }
 
+func (r *registryConfigurator) restoreUncleanShutdownDNS(*netip.Addr) error {
+	if err := r.restoreHostDNS(); err != nil {
+		return fmt.Errorf("restoring dns via registry: %w", err)
+	}
+	return nil
+}
+
 func removeRegistryKeyFromDNSPolicyConfig(regKeyPath string) error {
 	k, err := registry.OpenKey(registry.LOCAL_MACHINE, regKeyPath, registry.QUERY_VALUE)
 	if err == nil {
-		k.Close()
+		defer closer(k)
 		err = registry.DeleteKey(registry.LOCAL_MACHINE, regKeyPath)
 		if err != nil {
-			return fmt.Errorf("unable to remove existing key from registry, key: HKEY_LOCAL_MACHINE\\%s, error: %s", regKeyPath, err)
+			return fmt.Errorf("unable to remove existing key from registry, key: HKEY_LOCAL_MACHINE\\%s, error: %w", regKeyPath, err)
 		}
 	}
 	return nil
 }
+
+func closer(closer io.Closer) {
+	if err := closer.Close(); err != nil {
+		log.Errorf("failed to close: %s", err)
+	}
+}

client/internal/dns/hosts_dns_holder.go

@@ -0,0 +1,63 @@
+package dns
+
+import (
+	"fmt"
+	"net/netip"
+	"sync"
+
+	log "github.com/sirupsen/logrus"
+)
+
+type hostsDNSHolder struct {
+	unprotectedDNSList map[string]struct{}
+	mutex              sync.RWMutex
+}
+
+func newHostsDNSHolder() *hostsDNSHolder {
+	return &hostsDNSHolder{
+		unprotectedDNSList: make(map[string]struct{}),
+	}
+}
+
+func (h *hostsDNSHolder) set(list []string) {
+	h.mutex.Lock()
+	h.unprotectedDNSList = make(map[string]struct{})
+	for _, dns := range list {
+		dnsAddr, err := h.normalizeAddress(dns)
+		if err != nil {
+			continue
+		}
+		h.unprotectedDNSList[dnsAddr] = struct{}{}
+	}
+	h.mutex.Unlock()
+}
+
+func (h *hostsDNSHolder) get() map[string]struct{} {
+	h.mutex.RLock()
+	l := h.unprotectedDNSList
+	h.mutex.RUnlock()
+	return l
+}
+
+//nolint:unused
+func (h *hostsDNSHolder) isContain(upstream string) bool {
+	h.mutex.RLock()
+	defer h.mutex.RUnlock()
+
+	_, ok := h.unprotectedDNSList[upstream]
+	return ok
+}
+
+func (h *hostsDNSHolder) normalizeAddress(addr string) (string, error) {
+	a, err := netip.ParseAddr(addr)
+	if err != nil {
+		log.Errorf("invalid upstream IP address: %s, error: %s", addr, err)
+		return "", err
+	}
+
+	if a.Is4() {
+		return fmt.Sprintf("%s:53", addr), nil
+	} else {
+		return fmt.Sprintf("[%s]:53", addr), nil
+	}
+}

client/internal/dns/local.go

@@ -31,6 +31,8 @@ func (d *localResolver) ServeDNS(w dns.ResponseWriter, r *dns.Msg) {
 	response := d.lookupRecord(r)
 	if response != nil {
 		replyMessage.Answer = append(replyMessage.Answer, response)
+	} else {
+		replyMessage.Rcode = dns.RcodeNameError
 	}
 
 	err := w.WriteMsg(replyMessage)
@@ -52,7 +54,7 @@ func (d *localResolver) lookupRecord(r *dns.Msg) dns.RR {
 func (d *localResolver) registerRecord(record nbdns.SimpleRecord) error {
 	fullRecord, err := dns.NewRR(record.String())
 	if err != nil {
-		return err
+		return fmt.Errorf("register record: %w", err)
 	}
 
 	fullRecord.Header().Rdlength = record.Len()

client/internal/dns/network_manager_linux.go

@@ -5,8 +5,10 @@ package dns
 import (
 	"context"
 	"encoding/binary"
+	"errors"
 	"fmt"
 	"net/netip"
+	"strings"
 	"time"
 
 	"github.com/godbus/dbus/v5"
@@ -41,9 +43,13 @@ const (
 	networkManagerDbusPrimaryDNSPriority       int32 = -500
 	networkManagerDbusWithMatchDomainPriority  int32 = 0
 	networkManagerDbusSearchDomainOnlyPriority int32 = 50
-	supportedNetworkManagerVersionConstraint         = ">= 1.16, < 1.28"
 )
 
+var supportedNetworkManagerVersionConstraints = []string{
+	">= 1.16, < 1.27",
+	">= 1.44, < 1.45",
+}
+
 type networkManagerDbusConfigurator struct {
 	dbusLinkObject dbus.ObjectPath
 	routingAll     bool
@@ -71,19 +77,19 @@ func (s networkManagerConnSettings) cleanDeprecatedSettings() {
 	}
 }
 
-func newNetworkManagerDbusConfigurator(wgInterface WGIface) (hostManager, error) {
+func newNetworkManagerDbusConfigurator(wgInterface string) (hostManager, error) {
 	obj, closeConn, err := getDbusObject(networkManagerDest, networkManagerDbusObjectNode)
 	if err != nil {
-		return nil, err
+		return nil, fmt.Errorf("get nm dbus: %w", err)
 	}
 	defer closeConn()
 	var s string
-	err = obj.Call(networkManagerDbusGetDeviceByIPIfaceMethod, dbusDefaultFlag, wgInterface.Name()).Store(&s)
+	err = obj.Call(networkManagerDbusGetDeviceByIPIfaceMethod, dbusDefaultFlag, wgInterface).Store(&s)
 	if err != nil {
-		return nil, err
+		return nil, fmt.Errorf("call: %w", err)
 	}
 
-	log.Debugf("got network manager dbus Link Object: %s from net interface %s", s, wgInterface.Name())
+	log.Debugf("got network manager dbus Link Object: %s from net interface %s", s, wgInterface)
 
 	return &networkManagerDbusConfigurator{
 		dbusLinkObject: dbus.ObjectPath(s),
@@ -97,14 +103,14 @@ func (n *networkManagerDbusConfigurator) supportCustomPort() bool {
 func (n *networkManagerDbusConfigurator) applyDNSConfig(config HostDNSConfig) error {
 	connSettings, configVersion, err := n.getAppliedConnectionSettings()
 	if err != nil {
-		return fmt.Errorf("got an error while retrieving the applied connection settings, error: %s", err)
+		return fmt.Errorf("retrieving the applied connection settings, error: %w", err)
 	}
 
 	connSettings.cleanDeprecatedSettings()
 
 	dnsIP, err := netip.ParseAddr(config.ServerIP)
 	if err != nil {
-		return fmt.Errorf("unable to parse ip address, error: %s", err)
+		return fmt.Errorf("unable to parse ip address, error: %w", err)
 	}
 	convDNSIP := binary.LittleEndian.Uint32(dnsIP.AsSlice())
 	connSettings[networkManagerDbusIPv4Key][networkManagerDbusDNSKey] = dbus.MakeVariant([]uint32{convDNSIP})
@@ -145,23 +151,37 @@ func (n *networkManagerDbusConfigurator) applyDNSConfig(config HostDNSConfig) er
 	connSettings[networkManagerDbusIPv4Key][networkManagerDbusDNSPriorityKey] = dbus.MakeVariant(priority)
 	connSettings[networkManagerDbusIPv4Key][networkManagerDbusDNSSearchKey] = dbus.MakeVariant(newDomainList)
 
+	// create a backup for unclean shutdown detection before adding domains, as these might end up in the resolv.conf file.
+	// The file content itself is not important for network-manager restoration
+	if err := createUncleanShutdownIndicator(defaultResolvConfPath, networkManager, dnsIP.String()); err != nil {
+		log.Errorf("failed to create unclean shutdown resolv.conf backup: %s", err)
+	}
+
 	log.Infof("adding %d search domains and %d match domains. Search list: %s , Match list: %s", len(searchDomains), len(matchDomains), searchDomains, matchDomains)
 	err = n.reApplyConnectionSettings(connSettings, configVersion)
 	if err != nil {
-		return fmt.Errorf("got an error while reapplying the connection with new settings, error: %s", err)
+		return fmt.Errorf("reapplying the connection with new settings, error: %w", err)
 	}
 	return nil
 }
 
 func (n *networkManagerDbusConfigurator) restoreHostDNS() error {
 	// once the interface is gone network manager cleans all config associated with it
-	return n.deleteConnectionSettings()
+	if err := n.deleteConnectionSettings(); err != nil {
+		return fmt.Errorf("delete connection settings: %w", err)
+	}
+
+	if err := removeUncleanShutdownIndicator(); err != nil {
+		log.Errorf("failed to remove unclean shutdown resolv.conf backup: %s", err)
+	}
+
+	return nil
 }
 
 func (n *networkManagerDbusConfigurator) getAppliedConnectionSettings() (networkManagerConnSettings, networkManagerConfigVersion, error) {
 	obj, closeConn, err := getDbusObject(networkManagerDest, n.dbusLinkObject)
 	if err != nil {
-		return nil, 0, fmt.Errorf("got error while attempting to retrieve the applied connection settings, err: %s", err)
+		return nil, 0, fmt.Errorf("attempting to retrieve the applied connection settings, err: %w", err)
 	}
 	defer closeConn()
 
@@ -176,7 +196,7 @@ func (n *networkManagerDbusConfigurator) getAppliedConnectionSettings() (network
 	err = obj.CallWithContext(ctx, networkManagerDbusDeviceGetAppliedConnectionMethod, dbusDefaultFlag,
 		networkManagerDbusDefaultBehaviorFlag).Store(&connSettings, &configVersion)
 	if err != nil {
-		return nil, 0, fmt.Errorf("got error while calling GetAppliedConnection method with context, err: %s", err)
+		return nil, 0, fmt.Errorf("calling GetAppliedConnection method with context, err: %w", err)
 	}
 
 	return connSettings, configVersion, nil
@@ -185,7 +205,7 @@ func (n *networkManagerDbusConfigurator) getAppliedConnectionSettings() (network
 func (n *networkManagerDbusConfigurator) reApplyConnectionSettings(connSettings networkManagerConnSettings, configVersion networkManagerConfigVersion) error {
 	obj, closeConn, err := getDbusObject(networkManagerDest, n.dbusLinkObject)
 	if err != nil {
-		return fmt.Errorf("got error while attempting to retrieve the applied connection settings, err: %s", err)
+		return fmt.Errorf("attempting to retrieve the applied connection settings, err: %w", err)
 	}
 	defer closeConn()
 
@@ -195,7 +215,7 @@ func (n *networkManagerDbusConfigurator) reApplyConnectionSettings(connSettings
 	err = obj.CallWithContext(ctx, networkManagerDbusDeviceReapplyMethod, dbusDefaultFlag,
 		connSettings, configVersion, networkManagerDbusDefaultBehaviorFlag).Store()
 	if err != nil {
-		return fmt.Errorf("got error while calling ReApply method with context, err: %s", err)
+		return fmt.Errorf("calling ReApply method with context, err: %w", err)
 	}
 
 	return nil
@@ -204,21 +224,34 @@ func (n *networkManagerDbusConfigurator) reApplyConnectionSettings(connSettings
 func (n *networkManagerDbusConfigurator) deleteConnectionSettings() error {
 	obj, closeConn, err := getDbusObject(networkManagerDest, n.dbusLinkObject)
 	if err != nil {
-		return fmt.Errorf("got error while attempting to retrieve the applied connection settings, err: %s", err)
+		return fmt.Errorf("attempting to retrieve the applied connection settings, err: %w", err)
 	}
 	defer closeConn()
 
 	ctx, cancel := context.WithTimeout(context.TODO(), 5*time.Second)
 	defer cancel()
 
+	// this call is required to remove the device for DNS cleanup, even if it fails
 	err = obj.CallWithContext(ctx, networkManagerDbusDeviceDeleteMethod, dbusDefaultFlag).Store()
 	if err != nil {
-		return fmt.Errorf("got error while calling delete method with context, err: %s", err)
+		var dbusErr dbus.Error
+		if errors.As(err, &dbusErr) && dbusErr.Name == dbus.ErrMsgUnknownMethod.Name {
+			// interface is gone already
+			return nil
+		}
+		return fmt.Errorf("calling delete method with context, err: %s", err)
 	}
 
 	return nil
 }
 
+func (n *networkManagerDbusConfigurator) restoreUncleanShutdownDNS(*netip.Addr) error {
+	if err := n.restoreHostDNS(); err != nil {
+		return fmt.Errorf("restoring dns via network-manager: %w", err)
+	}
+	return nil
+}
+
 func isNetworkManagerSupported() bool {
 	return isNetworkManagerSupportedVersion() && isNetworkManagerSupportedMode()
 }
@@ -250,13 +283,13 @@ func isNetworkManagerSupportedMode() bool {
 func getNetworkManagerDNSProperty(property string, store any) error {
 	obj, closeConn, err := getDbusObject(networkManagerDest, networkManagerDbusDNSManagerObjectNode)
 	if err != nil {
-		return fmt.Errorf("got error while attempting to retrieve the network manager dns manager object, error: %s", err)
+		return fmt.Errorf("attempting to retrieve the network manager dns manager object, error: %w", err)
 	}
 	defer closeConn()
 
 	v, e := obj.GetProperty(property)
 	if e != nil {
-		return fmt.Errorf("got an error getting property %s: %v", property, e)
+		return fmt.Errorf("getting property %s: %w", property, e)
 	}
 
 	return v.Store(store)
@@ -278,15 +311,26 @@ func isNetworkManagerSupportedVersion() bool {
 	}
 	versionValue, err := parseVersion(value.Value().(string))
 	if err != nil {
+		log.Errorf("nm: parse version: %s", err)
 		return false
 	}
 
-	constraints, err := version.NewConstraint(supportedNetworkManagerVersionConstraint)
-	if err != nil {
-		return false
+	var supported bool
+	for _, constraint := range supportedNetworkManagerVersionConstraints {
+		constr, err := version.NewConstraint(constraint)
+		if err != nil {
+			log.Errorf("nm: create constraint: %s", err)
+			return false
+		}
+
+		if met := constr.Check(versionValue); met {
+			supported = true
+			break
+		}
 	}
 
-	return constraints.Check(versionValue)
+	log.Debugf("network manager constraints [%s] met: %t", strings.Join(supportedNetworkManagerVersionConstraints, " | "), supported)
+	return supported
 }
 
 func parseVersion(inputVersion string) (*version.Version, error) {

client/internal/dns/resolvconf_linux.go

@@ -5,6 +5,7 @@ package dns
 import (
 	"bytes"
 	"fmt"
+	"net/netip"
 	"os/exec"
 
 	log "github.com/sirupsen/logrus"
@@ -21,14 +22,14 @@ type resolvconf struct {
 }
 
 // supported "openresolv" only
-func newResolvConfConfigurator(wgInterface WGIface) (hostManager, error) {
+func newResolvConfConfigurator(wgInterface string) (hostManager, error) {
 	resolvConfEntries, err := parseDefaultResolvConf()
 	if err != nil {
-		log.Error(err)
+		log.Errorf("could not read original search domains from %s: %s", defaultResolvConfPath, err)
 	}
 
 	return &resolvconf{
-		ifaceName:             wgInterface.Name(),
+		ifaceName:             wgInterface,
 		originalSearchDomains: resolvConfEntries.searchDomains,
 		originalNameServers:   resolvConfEntries.nameServers,
 		othersConfigs:         resolvConfEntries.others,
@@ -44,7 +45,7 @@ func (r *resolvconf) applyDNSConfig(config HostDNSConfig) error {
 	if !config.RouteAll {
 		err = r.restoreHostDNS()
 		if err != nil {
-			log.Error(err)
+			log.Errorf("restore host dns: %s", err)
 		}
 		return fmt.Errorf("unable to configure DNS for this peer using resolvconf manager without a nameserver group with all domains configured")
 	}
@@ -52,14 +53,21 @@ func (r *resolvconf) applyDNSConfig(config HostDNSConfig) error {
 	searchDomainList := searchDomains(config)
 	searchDomainList = mergeSearchDomains(searchDomainList, r.originalSearchDomains)
 
+	options := prepareOptionsWithTimeout(r.othersConfigs, int(dnsFailoverTimeout.Seconds()), dnsFailoverAttempts)
+
 	buf := prepareResolvConfContent(
 		searchDomainList,
 		append([]string{config.ServerIP}, r.originalNameServers...),
-		r.othersConfigs)
+		options)
+
+	// create a backup for unclean shutdown detection before the resolv.conf is changed
+	if err := createUncleanShutdownIndicator(defaultResolvConfPath, resolvConfManager, config.ServerIP); err != nil {
+		log.Errorf("failed to create unclean shutdown resolv.conf backup: %s", err)
+	}
 
 	err = r.applyConfig(buf)
 	if err != nil {
-		return err
+		return fmt.Errorf("apply config: %w", err)
 	}
 
 	log.Infof("added %d search domains. Search list: %s", len(searchDomainList), searchDomainList)
@@ -67,20 +75,34 @@ func (r *resolvconf) applyDNSConfig(config HostDNSConfig) error {
 }
 
 func (r *resolvconf) restoreHostDNS() error {
+	// openresolv only, debian resolvconf doesn't support "-f"
 	cmd := exec.Command(resolvconfCommand, "-f", "-d", r.ifaceName)
 	_, err := cmd.Output()
 	if err != nil {
-		return fmt.Errorf("got an error while removing resolvconf configuration for %s interface, error: %s", r.ifaceName, err)
+		return fmt.Errorf("removing resolvconf configuration for %s interface, error: %w", r.ifaceName, err)
 	}
+
+	if err := removeUncleanShutdownIndicator(); err != nil {
+		log.Errorf("failed to remove unclean shutdown resolv.conf backup: %s", err)
+	}
+
 	return nil
 }
 
 func (r *resolvconf) applyConfig(content bytes.Buffer) error {
+	// openresolv only, debian resolvconf doesn't support "-x"
 	cmd := exec.Command(resolvconfCommand, "-x", "-a", r.ifaceName)
 	cmd.Stdin = &content
 	_, err := cmd.Output()
 	if err != nil {
-		return fmt.Errorf("got an error while applying resolvconf configuration for %s interface, error: %s", r.ifaceName, err)
+		return fmt.Errorf("applying resolvconf configuration for %s interface, error: %w", r.ifaceName, err)
+	}
+	return nil
+}
+
+func (r *resolvconf) restoreUncleanShutdownDNS(*netip.Addr) error {
+	if err := r.restoreHostDNS(); err != nil {
+		return fmt.Errorf("restoring dns for interface %s: %w", r.ifaceName, err)
 	}
 	return nil
 }

client/internal/dns/response_writer.go

@@ -31,10 +31,13 @@ func (r *responseWriter) RemoteAddr() net.Addr {
 func (r *responseWriter) WriteMsg(msg *dns.Msg) error {
 	buff, err := msg.Pack()
 	if err != nil {
-		return err
+		return fmt.Errorf("pack: %w", err)
 	}
-	_, err = r.Write(buff)
-	return err
+
+	if _, err := r.Write(buff); err != nil {
+		return fmt.Errorf("write: %w", err)
+	}
+	return nil
 }
 
 // Write writes a raw buffer back to the client.

client/internal/dns/server.go

@@ -4,6 +4,8 @@ import (
 	"context"
 	"fmt"
 	"net/netip"
+	"runtime"
+	"strings"
 	"sync"
 
 	"github.com/miekg/dns"
@@ -11,6 +13,7 @@ import (
 	log "github.com/sirupsen/logrus"
 
 	"github.com/netbirdio/netbird/client/internal/listener"
+	"github.com/netbirdio/netbird/client/internal/peer"
 	nbdns "github.com/netbirdio/netbird/dns"
 )
 
@@ -52,13 +55,14 @@ type DefaultServer struct {
 	currentConfig      HostDNSConfig
 
 	// permanent related properties
-	permanent        bool
-	hostsDnsList     []string
-	hostsDnsListLock sync.Mutex
+	permanent      bool
+	hostsDNSHolder *hostsDNSHolder
 
 	// make sense on mobile only
 	searchDomainNotifier *notifier
 	iosDnsManager        IosDnsManager
+
+	statusRecorder *peer.Status
 }
 
 type handlerWithStop interface {
@@ -73,7 +77,12 @@ type muxUpdate struct {
 }
 
 // NewDefaultServer returns a new dns server
-func NewDefaultServer(ctx context.Context, wgInterface WGIface, customAddress string) (*DefaultServer, error) {
+func NewDefaultServer(
+	ctx context.Context,
+	wgInterface WGIface,
+	customAddress string,
+	statusRecorder *peer.Status,
+) (*DefaultServer, error) {
 	var addrPort *netip.AddrPort
 	if customAddress != "" {
 		parsedAddrPort, err := netip.ParseAddrPort(customAddress)
@@ -90,15 +99,22 @@ func NewDefaultServer(ctx context.Context, wgInterface WGIface, customAddress st
 		dnsService = newServiceViaListener(wgInterface, addrPort)
 	}
 
-	return newDefaultServer(ctx, wgInterface, dnsService), nil
+	return newDefaultServer(ctx, wgInterface, dnsService, statusRecorder), nil
 }
 
 // NewDefaultServerPermanentUpstream returns a new dns server. It optimized for mobile systems
-func NewDefaultServerPermanentUpstream(ctx context.Context, wgInterface WGIface, hostsDnsList []string, config nbdns.Config, listener listener.NetworkChangeListener) *DefaultServer {
+func NewDefaultServerPermanentUpstream(
+	ctx context.Context,
+	wgInterface WGIface,
+	hostsDnsList []string,
+	config nbdns.Config,
+	listener listener.NetworkChangeListener,
+	statusRecorder *peer.Status,
+) *DefaultServer {
 	log.Debugf("host dns address list is: %v", hostsDnsList)
-	ds := newDefaultServer(ctx, wgInterface, newServiceViaMemory(wgInterface))
+	ds := newDefaultServer(ctx, wgInterface, newServiceViaMemory(wgInterface), statusRecorder)
+	ds.hostsDNSHolder.set(hostsDnsList)
 	ds.permanent = true
-	ds.hostsDnsList = hostsDnsList
 	ds.addHostRootZone()
 	ds.currentConfig = dnsConfigToHostDNSConfig(config, ds.service.RuntimeIP(), ds.service.RuntimePort())
 	ds.searchDomainNotifier = newNotifier(ds.SearchDomains())
@@ -108,13 +124,18 @@ func NewDefaultServerPermanentUpstream(ctx context.Context, wgInterface WGIface,
 }
 
 // NewDefaultServerIos returns a new dns server. It optimized for ios
-func NewDefaultServerIos(ctx context.Context, wgInterface WGIface, iosDnsManager IosDnsManager) *DefaultServer {
-	ds := newDefaultServer(ctx, wgInterface, newServiceViaMemory(wgInterface))
+func NewDefaultServerIos(
+	ctx context.Context,
+	wgInterface WGIface,
+	iosDnsManager IosDnsManager,
+	statusRecorder *peer.Status,
+) *DefaultServer {
+	ds := newDefaultServer(ctx, wgInterface, newServiceViaMemory(wgInterface), statusRecorder)
 	ds.iosDnsManager = iosDnsManager
 	return ds
 }
 
-func newDefaultServer(ctx context.Context, wgInterface WGIface, dnsService service) *DefaultServer {
+func newDefaultServer(ctx context.Context, wgInterface WGIface, dnsService service, statusRecorder *peer.Status) *DefaultServer {
 	ctx, stop := context.WithCancel(ctx)
 	defaultServer := &DefaultServer{
 		ctx:       ctx,
@@ -124,7 +145,9 @@ func newDefaultServer(ctx context.Context, wgInterface WGIface, dnsService servi
 		localResolver: &localResolver{
 			registeredMap: make(registrationMap),
 		},
-		wgInterface: wgInterface,
+		wgInterface:    wgInterface,
+		statusRecorder: statusRecorder,
+		hostsDNSHolder: newHostsDNSHolder(),
 	}
 
 	return defaultServer
@@ -142,12 +165,15 @@ func (s *DefaultServer) Initialize() (err error) {
 	if s.permanent {
 		err = s.service.Listen()
 		if err != nil {
-			return err
+			return fmt.Errorf("service listen: %w", err)
 		}
 	}
 
 	s.hostManager, err = s.initialize()
-	return err
+	if err != nil {
+		return fmt.Errorf("initialize: %w", err)
+	}
+	return nil
 }
 
 // DnsIP returns the DNS resolver server IP address
@@ -177,10 +203,8 @@ func (s *DefaultServer) Stop() {
 // OnUpdatedHostDNSServer update the DNS servers addresses for root zones
 // It will be applied if the mgm server do not enforce DNS settings for root zone
 func (s *DefaultServer) OnUpdatedHostDNSServer(hostsDnsList []string) {
-	s.hostsDnsListLock.Lock()
-	defer s.hostsDnsListLock.Unlock()
+	s.hostsDNSHolder.set(hostsDnsList)
 
-	s.hostsDnsList = hostsDnsList
 	_, ok := s.dnsMuxMap[nbdns.RootZone]
 	if ok {
 		log.Debugf("on new host DNS config but skip to apply it")
@@ -225,7 +249,7 @@ func (s *DefaultServer) UpdateDNSServer(serial uint64, update nbdns.Config) erro
 		}
 
 		if err := s.applyConfiguration(update); err != nil {
-			return err
+			return fmt.Errorf("apply configuration: %w", err)
 		}
 
 		s.updateSerial = serial
@@ -253,9 +277,15 @@ func (s *DefaultServer) SearchDomains() []string {
 // ProbeAvailability tests each upstream group's servers for availability
 // and deactivates the group if no server responds
 func (s *DefaultServer) ProbeAvailability() {
+	var wg sync.WaitGroup
 	for _, mux := range s.dnsMuxMap {
-		mux.probeAvailability()
+		wg.Add(1)
+		go func(mux handlerWithStop) {
+			defer wg.Done()
+			mux.probeAvailability()
+		}(mux)
 	}
+	wg.Wait()
 }
 
 func (s *DefaultServer) applyConfiguration(update nbdns.Config) error {
@@ -296,6 +326,8 @@ func (s *DefaultServer) applyConfiguration(update nbdns.Config) error {
 		s.searchDomainNotifier.onNewSearchDomains(s.SearchDomains())
 	}
 
+	s.updateNSGroupStates(update.NameServerGroups)
+
 	return nil
 }
 
@@ -335,7 +367,14 @@ func (s *DefaultServer) buildUpstreamHandlerUpdate(nameServerGroups []*nbdns.Nam
 			continue
 		}
 
-		handler, err := newUpstreamResolver(s.ctx, s.wgInterface.Name(), s.wgInterface.Address().IP, s.wgInterface.Address().Network)
+		handler, err := newUpstreamResolver(
+			s.ctx,
+			s.wgInterface.Name(),
+			s.wgInterface.Address().IP,
+			s.wgInterface.Address().Network,
+			s.statusRecorder,
+			s.hostsDNSHolder,
+		)
 		if err != nil {
 			return nil, fmt.Errorf("unable to create a new upstream resolver, error: %v", err)
 		}
@@ -413,9 +452,7 @@ func (s *DefaultServer) updateMux(muxUpdates []muxUpdate) {
 		_, found := muxUpdateMap[key]
 		if !found {
 			if !isContainRootUpdate && key == nbdns.RootZone {
-				s.hostsDnsListLock.Lock()
 				s.addHostRootZone()
-				s.hostsDnsListLock.Unlock()
 				existingHandler.stop()
 			} else {
 				existingHandler.stop()
@@ -457,14 +494,14 @@ func getNSHostPort(ns nbdns.NameServer) string {
 func (s *DefaultServer) upstreamCallbacks(
 	nsGroup *nbdns.NameServerGroup,
 	handler dns.Handler,
-) (deactivate func(), reactivate func()) {
+) (deactivate func(error), reactivate func()) {
 	var removeIndex map[string]int
-	deactivate = func() {
+	deactivate = func(err error) {
 		s.mux.Lock()
 		defer s.mux.Unlock()
 
 		l := log.WithField("nameservers", nsGroup.NameServers)
-		l.Info("temporary deactivate nameservers group due timeout")
+		l.Info("Temporarily deactivating nameservers group due to timeout")
 
 		removeIndex = make(map[string]int)
 		for _, domain := range nsGroup.Domains {
@@ -473,6 +510,7 @@ func (s *DefaultServer) upstreamCallbacks(
 		if nsGroup.Primary {
 			removeIndex[nbdns.RootZone] = -1
 			s.currentConfig.RouteAll = false
+			s.service.DeregisterMux(nbdns.RootZone)
 		}
 
 		for i, item := range s.currentConfig.Domains {
@@ -482,9 +520,17 @@ func (s *DefaultServer) upstreamCallbacks(
 				removeIndex[item.Domain] = i
 			}
 		}
+
 		if err := s.hostManager.applyDNSConfig(s.currentConfig); err != nil {
-			l.WithError(err).Error("fail to apply nameserver deactivation on the host")
+			l.Errorf("Failed to apply nameserver deactivation on the host: %v", err)
 		}
+
+		if runtime.GOOS == "android" && nsGroup.Primary && len(s.hostsDNSHolder.get()) > 0 {
+			s.addHostRootZone()
+		}
+
+		s.updateNSState(nsGroup, err, false)
+
 	}
 	reactivate = func() {
 		s.mux.Lock()
@@ -503,36 +549,79 @@ func (s *DefaultServer) upstreamCallbacks(
 
 		if nsGroup.Primary {
 			s.currentConfig.RouteAll = true
+			s.service.RegisterMux(nbdns.RootZone, handler)
 		}
 		if err := s.hostManager.applyDNSConfig(s.currentConfig); err != nil {
 			l.WithError(err).Error("reactivate temporary disabled nameserver group, DNS update apply")
 		}
+
+		s.updateNSState(nsGroup, nil, true)
 	}
 	return
 }
 
 func (s *DefaultServer) addHostRootZone() {
-	handler, err := newUpstreamResolver(s.ctx, s.wgInterface.Name(), s.wgInterface.Address().IP, s.wgInterface.Address().Network)
+	handler, err := newUpstreamResolver(
+		s.ctx,
+		s.wgInterface.Name(),
+		s.wgInterface.Address().IP,
+		s.wgInterface.Address().Network,
+		s.statusRecorder,
+		s.hostsDNSHolder,
+	)
 	if err != nil {
 		log.Errorf("unable to create a new upstream resolver, error: %v", err)
 		return
 	}
-	handler.upstreamServers = make([]string, len(s.hostsDnsList))
-	for n, ua := range s.hostsDnsList {
-		a, err := netip.ParseAddr(ua)
-		if err != nil {
-			log.Errorf("invalid upstream IP address: %s, error: %s", ua, err)
-			continue
-		}
 
-		ipString := ua
-		if !a.Is4() {
-			ipString = fmt.Sprintf("[%s]", ua)
-		}
-
-		handler.upstreamServers[n] = fmt.Sprintf("%s:53", ipString)
+	handler.upstreamServers = make([]string, 0)
+	for k := range s.hostsDNSHolder.get() {
+		handler.upstreamServers = append(handler.upstreamServers, k)
 	}
-	handler.deactivate = func() {}
+	handler.deactivate = func(error) {}
 	handler.reactivate = func() {}
 	s.service.RegisterMux(nbdns.RootZone, handler)
 }
+
+func (s *DefaultServer) updateNSGroupStates(groups []*nbdns.NameServerGroup) {
+	var states []peer.NSGroupState
+
+	for _, group := range groups {
+		var servers []string
+		for _, ns := range group.NameServers {
+			servers = append(servers, fmt.Sprintf("%s:%d", ns.IP, ns.Port))
+		}
+
+		state := peer.NSGroupState{
+			ID:      generateGroupKey(group),
+			Servers: servers,
+			Domains: group.Domains,
+			// The probe will determine the state, default enabled
+			Enabled: true,
+			Error:   nil,
+		}
+		states = append(states, state)
+	}
+	s.statusRecorder.UpdateDNSStates(states)
+}
+
+func (s *DefaultServer) updateNSState(nsGroup *nbdns.NameServerGroup, err error, enabled bool) {
+	states := s.statusRecorder.GetDNSStates()
+	id := generateGroupKey(nsGroup)
+	for i, state := range states {
+		if state.ID == id {
+			states[i].Enabled = enabled
+			states[i].Error = err
+			break
+		}
+	}
+	s.statusRecorder.UpdateDNSStates(states)
+}
+
+func generateGroupKey(nsGroup *nbdns.NameServerGroup) string {
+	var servers []string
+	for _, ns := range nsGroup.NameServers {
+		servers = append(servers, fmt.Sprintf("%s:%d", ns.IP, ns.Port))
+	}
+	return fmt.Sprintf("%s_%s_%s", nsGroup.ID, nsGroup.Name, strings.Join(servers, ","))
+}

client/internal/dns/server_android.go

@@ -1,5 +1,5 @@
 package dns
 
 func (s *DefaultServer) initialize() (manager hostManager, err error) {
-	return newHostManager(s.wgInterface)
+	return newHostManager()
 }

client/internal/dns/server_darwin.go

@@ -3,5 +3,5 @@
 package dns
 
 func (s *DefaultServer) initialize() (manager hostManager, err error) {
-	return newHostManager(s.wgInterface)
+	return newHostManager()
 }

client/internal/dns/server_linux.go

@@ -3,5 +3,5 @@
 package dns
 
 func (s *DefaultServer) initialize() (manager hostManager, err error) {
-	return newHostManager(s.wgInterface)
+	return newHostManager(s.wgInterface.Name())
 }

client/internal/dns/server_test.go

@@ -15,6 +15,7 @@ import (
 	"golang.zx2c4.com/wireguard/wgctrl/wgtypes"
 
 	"github.com/netbirdio/netbird/client/firewall/uspfilter"
+	"github.com/netbirdio/netbird/client/internal/peer"
 	"github.com/netbirdio/netbird/client/internal/stdnet"
 	nbdns "github.com/netbirdio/netbird/dns"
 	"github.com/netbirdio/netbird/formatter"
@@ -274,7 +275,7 @@ func TestUpdateDNSServer(t *testing.T) {
 					t.Log(err)
 				}
 			}()
-			dnsServer, err := NewDefaultServer(context.Background(), wgIface, "")
+			dnsServer, err := NewDefaultServer(context.Background(), wgIface, "", &peer.Status{})
 			if err != nil {
 				t.Fatal(err)
 			}
@@ -375,7 +376,7 @@ func TestDNSFakeResolverHandleUpdates(t *testing.T) {
 		return
 	}
 
-	dnsServer, err := NewDefaultServer(context.Background(), wgIface, "")
+	dnsServer, err := NewDefaultServer(context.Background(), wgIface, "", &peer.Status{})
 	if err != nil {
 		t.Errorf("create DNS server: %v", err)
 		return
@@ -470,7 +471,7 @@ func TestDNSServerStartStop(t *testing.T) {
 
 	for _, testCase := range testCases {
 		t.Run(testCase.name, func(t *testing.T) {
-			dnsServer, err := NewDefaultServer(context.Background(), &mocWGIface{}, testCase.addrPort)
+			dnsServer, err := NewDefaultServer(context.Background(), &mocWGIface{}, testCase.addrPort, &peer.Status{})
 			if err != nil {
 				t.Fatalf("%v", err)
 			}
@@ -541,6 +542,7 @@ func TestDNSServerUpstreamDeactivateCallback(t *testing.T) {
 				{false, "domain2", false},
 			},
 		},
+		statusRecorder: &peer.Status{},
 	}
 
 	var domainsUpdate string
@@ -563,7 +565,7 @@ func TestDNSServerUpstreamDeactivateCallback(t *testing.T) {
 		},
 	}, nil)
 
-	deactivate()
+	deactivate(nil)
 	expected := "domain0,domain2"
 	domains := []string{}
 	for _, item := range server.currentConfig.Domains {
@@ -601,7 +603,7 @@ func TestDNSPermanent_updateHostDNS_emptyUpstream(t *testing.T) {
 
 	var dnsList []string
 	dnsConfig := nbdns.Config{}
-	dnsServer := NewDefaultServerPermanentUpstream(context.Background(), wgIFace, dnsList, dnsConfig, nil)
+	dnsServer := NewDefaultServerPermanentUpstream(context.Background(), wgIFace, dnsList, dnsConfig, nil, &peer.Status{})
 	err = dnsServer.Initialize()
 	if err != nil {
 		t.Errorf("failed to initialize DNS server: %v", err)
@@ -625,7 +627,7 @@ func TestDNSPermanent_updateUpstream(t *testing.T) {
 	}
 	defer wgIFace.Close()
 	dnsConfig := nbdns.Config{}
-	dnsServer := NewDefaultServerPermanentUpstream(context.Background(), wgIFace, []string{"8.8.8.8"}, dnsConfig, nil)
+	dnsServer := NewDefaultServerPermanentUpstream(context.Background(), wgIFace, []string{"8.8.8.8"}, dnsConfig, nil, &peer.Status{})
 	err = dnsServer.Initialize()
 	if err != nil {
 		t.Errorf("failed to initialize DNS server: %v", err)
@@ -717,7 +719,7 @@ func TestDNSPermanent_matchOnly(t *testing.T) {
 	}
 	defer wgIFace.Close()
 	dnsConfig := nbdns.Config{}
-	dnsServer := NewDefaultServerPermanentUpstream(context.Background(), wgIFace, []string{"8.8.8.8"}, dnsConfig, nil)
+	dnsServer := NewDefaultServerPermanentUpstream(context.Background(), wgIFace, []string{"8.8.8.8"}, dnsConfig, nil, &peer.Status{})
 	err = dnsServer.Initialize()
 	if err != nil {
 		t.Errorf("failed to initialize DNS server: %v", err)
@@ -748,6 +750,11 @@ func TestDNSPermanent_matchOnly(t *testing.T) {
 						NSType: nbdns.UDPNameServerType,
 						Port:   53,
 					},
+					{
+						IP:     netip.MustParseAddr("9.9.9.9"),
+						NSType: nbdns.UDPNameServerType,
+						Port:   53,
+					},
 				},
 				Domains: []string{"customdomain.com"},
 				Primary: false,

client/internal/dns/service_listener.go

@@ -28,7 +28,7 @@ type serviceViaListener struct {
 	customAddr        *netip.AddrPort
 	server            *dns.Server
 	listenIP          string
-	listenPort        int
+	listenPort        uint16
 	listenerIsRunning bool
 	listenerFlagLock  sync.Mutex
 	ebpfService       ebpfMgr.Manager
@@ -63,18 +63,9 @@ func (s *serviceViaListener) Listen() error {
 	s.listenIP, s.listenPort, err = s.evalListenAddress()
 	if err != nil {
 		log.Errorf("failed to eval runtime address: %s", err)
-		return err
+		return fmt.Errorf("eval listen address: %w", err)
 	}
 	s.server.Addr = fmt.Sprintf("%s:%d", s.listenIP, s.listenPort)
-
-	if s.shouldApplyPortFwd() {
-		s.ebpfService = ebpf.GetEbpfManagerInstance()
-		err = s.ebpfService.LoadDNSFwd(s.listenIP, s.listenPort)
-		if err != nil {
-			log.Warnf("failed to load DNS port forwarder, custom port may not work well on some Linux operating systems: %s", err)
-			s.ebpfService = nil
-		}
-	}
 	log.Debugf("starting dns on %s", s.server.Addr)
 	go func() {
 		s.setListenerStatus(true)
@@ -128,7 +119,7 @@ func (s *serviceViaListener) RuntimePort() int {
 	if s.ebpfService != nil {
 		return defaultPort
 	} else {
-		return s.listenPort
+		return int(s.listenPort)
 	}
 }
 
@@ -140,54 +131,112 @@ func (s *serviceViaListener) setListenerStatus(running bool) {
 	s.listenerIsRunning = running
 }
 
-func (s *serviceViaListener) getFirstListenerAvailable() (string, int, error) {
-	ips := []string{defaultIP, customIP}
+// evalListenAddress figure out the listen address for the DNS server
+// first check the 53 port availability on WG interface or lo, if not success
+// pick a random port on WG interface for eBPF, if not success
+// check the 5053 port availability on WG interface or lo without eBPF usage,
+func (s *serviceViaListener) evalListenAddress() (string, uint16, error) {
+	if s.customAddr != nil {
+		return s.customAddr.Addr().String(), s.customAddr.Port(), nil
+	}
+
+	ip, ok := s.testFreePort(defaultPort)
+	if ok {
+		return ip, defaultPort, nil
+	}
+
+	ebpfSrv, port, ok := s.tryToUseeBPF()
+	if ok {
+		s.ebpfService = ebpfSrv
+		return s.wgInterface.Address().IP.String(), port, nil
+	}
+
+	ip, ok = s.testFreePort(customPort)
+	if ok {
+		return ip, customPort, nil
+	}
+
+	return "", 0, fmt.Errorf("failed to find a free port for DNS server")
+}
+
+func (s *serviceViaListener) testFreePort(port int) (string, bool) {
+	var ips []string
 	if runtime.GOOS != "darwin" {
-		ips = append([]string{s.wgInterface.Address().IP.String()}, ips...)
+		ips = []string{s.wgInterface.Address().IP.String(), defaultIP, customIP}
+	} else {
+		ips = []string{defaultIP, customIP}
 	}
-	ports := []int{defaultPort, customPort}
-	for _, port := range ports {
-		for _, ip := range ips {
-			addrString := fmt.Sprintf("%s:%d", ip, port)
-			udpAddr := net.UDPAddrFromAddrPort(netip.MustParseAddrPort(addrString))
-			probeListener, err := net.ListenUDP("udp", udpAddr)
-			if err == nil {
-				err = probeListener.Close()
-				if err != nil {
-					log.Errorf("got an error closing the probe listener, error: %s", err)
-				}
-				return ip, port, nil
-			}
-			log.Warnf("binding dns on %s is not available, error: %s", addrString, err)
+
+	for _, ip := range ips {
+		if !s.tryToBind(ip, port) {
+			continue
 		}
+
+		return ip, true
 	}
-	return "", 0, fmt.Errorf("unable to find an unused ip and port combination. IPs tested: %v and ports %v", ips, ports)
+	return "", false
 }
 
-func (s *serviceViaListener) evalListenAddress() (string, int, error) {
-	if s.customAddr != nil {
-		return s.customAddr.Addr().String(), int(s.customAddr.Port()), nil
-	}
-
-	return s.getFirstListenerAvailable()
-}
-
-// shouldApplyPortFwd decides whether to apply eBPF program to capture DNS traffic on port 53.
-// This is needed because on some operating systems if we start a DNS server not on a default port 53, the domain name
-// resolution won't work.
-// So, in case we are running on Linux and picked a non-default port (53) we should fall back to the eBPF solution that will capture
-// traffic on port 53 and forward it to a local DNS server running on 5053.
-func (s *serviceViaListener) shouldApplyPortFwd() bool {
-	if runtime.GOOS != "linux" {
+func (s *serviceViaListener) tryToBind(ip string, port int) bool {
+	addrString := fmt.Sprintf("%s:%d", ip, port)
+	udpAddr := net.UDPAddrFromAddrPort(netip.MustParseAddrPort(addrString))
+	probeListener, err := net.ListenUDP("udp", udpAddr)
+	if err != nil {
+		log.Warnf("binding dns on %s is not available, error: %s", addrString, err)
 		return false
 	}
 
-	if s.customAddr != nil {
-		return false
-	}
-
-	if s.listenPort == defaultPort {
-		return false
+	err = probeListener.Close()
+	if err != nil {
+		log.Errorf("got an error closing the probe listener, error: %s", err)
 	}
 	return true
 }
+
+// tryToUseeBPF decides whether to apply eBPF program to capture DNS traffic on port 53.
+// This is needed because on some operating systems if we start a DNS server not on a default port 53,
+// the domain name  resolution won't work. So, in case we are running on Linux and picked a free
+// port we should fall back to the eBPF solution that will capture traffic on port 53 and forward
+// it to a local DNS server running on the chosen port.
+func (s *serviceViaListener) tryToUseeBPF() (ebpfMgr.Manager, uint16, bool) {
+	if runtime.GOOS != "linux" {
+		return nil, 0, false
+	}
+
+	port, err := s.generateFreePort() //nolint:staticcheck,unused
+	if err != nil {
+		log.Warnf("failed to generate a free port for eBPF DNS forwarder server: %s", err)
+		return nil, 0, false
+	}
+
+	ebpfSrv := ebpf.GetEbpfManagerInstance()
+	err = ebpfSrv.LoadDNSFwd(s.wgInterface.Address().IP.String(), int(port))
+	if err != nil {
+		log.Warnf("failed to load DNS forwarder eBPF program, error: %s", err)
+		return nil, 0, false
+	}
+
+	return ebpfSrv, port, true
+}
+
+func (s *serviceViaListener) generateFreePort() (uint16, error) {
+	ok := s.tryToBind(s.wgInterface.Address().IP.String(), customPort)
+	if ok {
+		return customPort, nil
+	}
+
+	udpAddr := net.UDPAddrFromAddrPort(netip.MustParseAddrPort("0.0.0.0:0"))
+	probeListener, err := net.ListenUDP("udp", udpAddr)
+	if err != nil {
+		log.Debugf("failed to bind random port for DNS: %s", err)
+		return 0, err
+	}
+
+	addrPort := netip.MustParseAddrPort(probeListener.LocalAddr().String()) // might panic if address is incorrect
+	err = probeListener.Close()
+	if err != nil {
+		log.Debugf("failed to free up DNS port: %s", err)
+		return 0, err
+	}
+	return addrPort.Port(), nil
+}

client/internal/dns/service_memory.go

@@ -44,7 +44,7 @@ func (s *serviceViaMemory) Listen() error {
 	var err error
 	s.udpFilterHookID, err = s.filterDNSTraffic()
 	if err != nil {
-		return err
+		return fmt.Errorf("filter dns traffice: %w", err)
 	}
 	s.listenerIsRunning = true
 

client/internal/dns/systemd_linux.go

@@ -4,6 +4,7 @@ package dns
 
 import (
 	"context"
+	"errors"
 	"fmt"
 	"net"
 	"net/netip"
@@ -30,6 +31,8 @@ const (
 	systemdDbusSetDefaultRouteMethodSuffix = systemdDbusLinkInterface + ".SetDefaultRoute"
 	systemdDbusSetDomainsMethodSuffix      = systemdDbusLinkInterface + ".SetDomains"
 	systemdDbusResolvConfModeForeign       = "foreign"
+
+	dbusErrorUnknownObject = "org.freedesktop.DBus.Error.UnknownObject"
 )
 
 type systemdDbusConfigurator struct {
@@ -52,22 +55,22 @@ type systemdDbusLinkDomainsInput struct {
 	MatchOnly bool
 }
 
-func newSystemdDbusConfigurator(wgInterface WGIface) (hostManager, error) {
-	iface, err := net.InterfaceByName(wgInterface.Name())
+func newSystemdDbusConfigurator(wgInterface string) (hostManager, error) {
+	iface, err := net.InterfaceByName(wgInterface)
 	if err != nil {
-		return nil, err
+		return nil, fmt.Errorf("get interface: %w", err)
 	}
 
 	obj, closeConn, err := getDbusObject(systemdResolvedDest, systemdDbusObjectNode)
 	if err != nil {
-		return nil, err
+		return nil, fmt.Errorf("get dbus resolved dest: %w", err)
 	}
 	defer closeConn()
 
 	var s string
 	err = obj.Call(systemdDbusGetLinkMethod, dbusDefaultFlag, iface.Index).Store(&s)
 	if err != nil {
-		return nil, err
+		return nil, fmt.Errorf("get dbus link method: %w", err)
 	}
 
 	log.Debugf("got dbus Link interface: %s from net interface %s and index %d", s, iface.Name, iface.Index)
@@ -84,7 +87,7 @@ func (s *systemdDbusConfigurator) supportCustomPort() bool {
 func (s *systemdDbusConfigurator) applyDNSConfig(config HostDNSConfig) error {
 	parsedIP, err := netip.ParseAddr(config.ServerIP)
 	if err != nil {
-		return fmt.Errorf("unable to parse ip address, error: %s", err)
+		return fmt.Errorf("unable to parse ip address, error: %w", err)
 	}
 	ipAs4 := parsedIP.As4()
 	defaultLinkInput := systemdDbusDNSInput{
@@ -93,7 +96,7 @@ func (s *systemdDbusConfigurator) applyDNSConfig(config HostDNSConfig) error {
 	}
 	err = s.callLinkMethod(systemdDbusSetDNSMethodSuffix, []systemdDbusDNSInput{defaultLinkInput})
 	if err != nil {
-		return fmt.Errorf("setting the interface DNS server %s:%d failed with error: %s", config.ServerIP, config.ServerPort, err)
+		return fmt.Errorf("setting the interface DNS server %s:%d failed with error: %w", config.ServerIP, config.ServerPort, err)
 	}
 
 	var (
@@ -121,7 +124,7 @@ func (s *systemdDbusConfigurator) applyDNSConfig(config HostDNSConfig) error {
 		log.Infof("configured %s:%d as main DNS forwarder for this peer", config.ServerIP, config.ServerPort)
 		err = s.callLinkMethod(systemdDbusSetDefaultRouteMethodSuffix, true)
 		if err != nil {
-			return fmt.Errorf("setting link as default dns router, failed with error: %s", err)
+			return fmt.Errorf("setting link as default dns router, failed with error: %w", err)
 		}
 		domainsInput = append(domainsInput, systemdDbusLinkDomainsInput{
 			Domain:    nbdns.RootZone,
@@ -132,6 +135,12 @@ func (s *systemdDbusConfigurator) applyDNSConfig(config HostDNSConfig) error {
 		log.Infof("removing %s:%d as main DNS forwarder for this peer", config.ServerIP, config.ServerPort)
 	}
 
+	// create a backup for unclean shutdown detection before adding domains, as these might end up in the resolv.conf file.
+	// The file content itself is not important for systemd restoration
+	if err := createUncleanShutdownIndicator(defaultResolvConfPath, systemdManager, parsedIP.String()); err != nil {
+		log.Errorf("failed to create unclean shutdown resolv.conf backup: %s", err)
+	}
+
 	log.Infof("adding %d search domains and %d match domains. Search list: %s , Match list: %s", len(searchDomains), len(matchDomains), searchDomains, matchDomains)
 	err = s.setDomainsForInterface(domainsInput)
 	if err != nil {
@@ -143,7 +152,7 @@ func (s *systemdDbusConfigurator) applyDNSConfig(config HostDNSConfig) error {
 func (s *systemdDbusConfigurator) setDomainsForInterface(domainsInput []systemdDbusLinkDomainsInput) error {
 	err := s.callLinkMethod(systemdDbusSetDomainsMethodSuffix, domainsInput)
 	if err != nil {
-		return fmt.Errorf("setting domains configuration failed with error: %s", err)
+		return fmt.Errorf("setting domains configuration failed with error: %w", err)
 	}
 	return s.flushCaches()
 }
@@ -153,17 +162,29 @@ func (s *systemdDbusConfigurator) restoreHostDNS() error {
 	if !isDbusListenerRunning(systemdResolvedDest, s.dbusLinkObject) {
 		return nil
 	}
+
+	// this call is required for DNS cleanup, even if it fails
 	err := s.callLinkMethod(systemdDbusRevertMethodSuffix, nil)
 	if err != nil {
-		return fmt.Errorf("unable to revert link configuration, got error: %s", err)
+		var dbusErr dbus.Error
+		if errors.As(err, &dbusErr) && dbusErr.Name == dbusErrorUnknownObject {
+			// interface is gone already
+			return nil
+		}
+		return fmt.Errorf("unable to revert link configuration, got error: %w", err)
 	}
+
+	if err := removeUncleanShutdownIndicator(); err != nil {
+		log.Errorf("failed to remove unclean shutdown resolv.conf backup: %s", err)
+	}
+
 	return s.flushCaches()
 }
 
 func (s *systemdDbusConfigurator) flushCaches() error {
 	obj, closeConn, err := getDbusObject(systemdResolvedDest, systemdDbusObjectNode)
 	if err != nil {
-		return fmt.Errorf("got error while attempting to retrieve the object %s, err: %s", systemdDbusObjectNode, err)
+		return fmt.Errorf("attempting to retrieve the object %s, err: %w", systemdDbusObjectNode, err)
 	}
 	defer closeConn()
 	ctx, cancel := context.WithTimeout(context.TODO(), 5*time.Second)
@@ -171,7 +192,7 @@ func (s *systemdDbusConfigurator) flushCaches() error {
 
 	err = obj.CallWithContext(ctx, systemdDbusFlushCachesMethod, dbusDefaultFlag).Store()
 	if err != nil {
-		return fmt.Errorf("got error while calling the FlushCaches method with context, err: %s", err)
+		return fmt.Errorf("calling the FlushCaches method with context, err: %w", err)
 	}
 
 	return nil
@@ -180,7 +201,7 @@ func (s *systemdDbusConfigurator) flushCaches() error {
 func (s *systemdDbusConfigurator) callLinkMethod(method string, value any) error {
 	obj, closeConn, err := getDbusObject(systemdResolvedDest, s.dbusLinkObject)
 	if err != nil {
-		return fmt.Errorf("got error while attempting to retrieve the object, err: %s", err)
+		return fmt.Errorf("attempting to retrieve the object, err: %w", err)
 	}
 	defer closeConn()
 
@@ -194,22 +215,29 @@ func (s *systemdDbusConfigurator) callLinkMethod(method string, value any) error
 	}
 
 	if err != nil {
-		return fmt.Errorf("got error while calling command with context, err: %s", err)
+		return fmt.Errorf("calling command with context, err: %w", err)
 	}
 
 	return nil
 }
 
+func (s *systemdDbusConfigurator) restoreUncleanShutdownDNS(*netip.Addr) error {
+	if err := s.restoreHostDNS(); err != nil {
+		return fmt.Errorf("restoring dns via systemd: %w", err)
+	}
+	return nil
+}
+
 func getSystemdDbusProperty(property string, store any) error {
 	obj, closeConn, err := getDbusObject(systemdResolvedDest, systemdDbusObjectNode)
 	if err != nil {
-		return fmt.Errorf("got error while attempting to retrieve the systemd dns manager object, error: %s", err)
+		return fmt.Errorf("attempting to retrieve the systemd dns manager object, error: %w", err)
 	}
 	defer closeConn()
 
 	v, e := obj.GetProperty(property)
 	if e != nil {
-		return fmt.Errorf("got an error getting property %s: %v", property, e)
+		return fmt.Errorf("getting property %s: %w", property, e)
 	}
 
 	return v.Store(store)

client/internal/dns/unclean_shutdown_android.go

@@ -0,0 +1,5 @@
+package dns
+
+func CheckUncleanShutdown(string) error {
+	return nil
+}

client/internal/dns/unclean_shutdown_darwin.go

@@ -0,0 +1,59 @@
+//go:build !ios
+
+package dns
+
+import (
+	"errors"
+	"fmt"
+	"io/fs"
+	"os"
+	"path/filepath"
+
+	log "github.com/sirupsen/logrus"
+)
+
+const fileUncleanShutdownFileLocation = "/var/lib/netbird/unclean_shutdown_dns"
+
+func CheckUncleanShutdown(string) error {
+	if _, err := os.Stat(fileUncleanShutdownFileLocation); err != nil {
+		if errors.Is(err, fs.ErrNotExist) {
+			// no file -> clean shutdown
+			return nil
+		} else {
+			return fmt.Errorf("state: %w", err)
+		}
+	}
+
+	log.Warnf("detected unclean shutdown, file %s exists. Restoring unclean shutdown dns settings.", fileUncleanShutdownFileLocation)
+
+	manager, err := newHostManager()
+	if err != nil {
+		return fmt.Errorf("create host manager: %w", err)
+	}
+
+	if err := manager.restoreUncleanShutdownDNS(nil); err != nil {
+		return fmt.Errorf("restore unclean shutdown backup: %w", err)
+	}
+
+	return nil
+}
+
+func createUncleanShutdownIndicator() error {
+	dir := filepath.Dir(fileUncleanShutdownFileLocation)
+	if err := os.MkdirAll(dir, os.FileMode(0755)); err != nil {
+		return fmt.Errorf("create dir %s: %w", dir, err)
+	}
+
+	if err := os.WriteFile(fileUncleanShutdownFileLocation, nil, 0644); err != nil { //nolint:gosec
+		return fmt.Errorf("create %s: %w", fileUncleanShutdownFileLocation, err)
+	}
+
+	return nil
+}
+
+func removeUncleanShutdownIndicator() error {
+	if err := os.Remove(fileUncleanShutdownFileLocation); err != nil && !errors.Is(err, fs.ErrNotExist) {
+		return fmt.Errorf("remove %s: %w", fileUncleanShutdownFileLocation, err)
+	}
+	return nil
+}

client/internal/dns/unclean_shutdown_ios.go

@@ -0,0 +1,5 @@
+package dns
+
+func CheckUncleanShutdown(string) error {
+	return nil
+}

client/internal/dns/unclean_shutdown_linux.go

@@ -0,0 +1,96 @@
+//go:build !android
+
+package dns
+
+import (
+	"errors"
+	"fmt"
+	"io/fs"
+	"net/netip"
+	"os"
+	"path/filepath"
+	"strings"
+
+	log "github.com/sirupsen/logrus"
+)
+
+const (
+	fileUncleanShutdownResolvConfLocation  = "/var/lib/netbird/resolv.conf"
+	fileUncleanShutdownManagerTypeLocation = "/var/lib/netbird/manager"
+)
+
+func CheckUncleanShutdown(wgIface string) error {
+	if _, err := os.Stat(fileUncleanShutdownResolvConfLocation); err != nil {
+		if errors.Is(err, fs.ErrNotExist) {
+			// no file -> clean shutdown
+			return nil
+		} else {
+			return fmt.Errorf("state: %w", err)
+		}
+	}
+
+	log.Warnf("detected unclean shutdown, file %s exists", fileUncleanShutdownResolvConfLocation)
+
+	managerData, err := os.ReadFile(fileUncleanShutdownManagerTypeLocation)
+	if err != nil {
+		return fmt.Errorf("read %s: %w", fileUncleanShutdownManagerTypeLocation, err)
+	}
+
+	managerFields := strings.Split(string(managerData), ",")
+	if len(managerFields) < 2 {
+		return errors.New("split manager data: insufficient number of fields")
+	}
+	osManagerTypeStr, dnsAddressStr := managerFields[0], managerFields[1]
+
+	dnsAddress, err := netip.ParseAddr(dnsAddressStr)
+	if err != nil {
+		return fmt.Errorf("parse dns address %s failed: %w", dnsAddressStr, err)
+	}
+
+	log.Warnf("restoring unclean shutdown dns settings via previously detected manager: %s", osManagerTypeStr)
+
+	// determine os manager type, so we can invoke the respective restore action
+	osManagerType, err := newOsManagerType(osManagerTypeStr)
+	if err != nil {
+		return fmt.Errorf("detect previous host manager: %w", err)
+	}
+
+	manager, err := newHostManagerFromType(wgIface, osManagerType)
+	if err != nil {
+		return fmt.Errorf("create previous host manager: %w", err)
+	}
+
+	if err := manager.restoreUncleanShutdownDNS(&dnsAddress); err != nil {
+		return fmt.Errorf("restore unclean shutdown backup: %w", err)
+	}
+
+	return nil
+}
+
+func createUncleanShutdownIndicator(sourcePath string, managerType osManagerType, dnsAddress string) error {
+	dir := filepath.Dir(fileUncleanShutdownResolvConfLocation)
+	if err := os.MkdirAll(dir, os.FileMode(0755)); err != nil {
+		return fmt.Errorf("create dir %s: %w", dir, err)
+	}
+
+	if err := copyFile(sourcePath, fileUncleanShutdownResolvConfLocation); err != nil {
+		return fmt.Errorf("create %s: %w", sourcePath, err)
+	}
+
+	managerData := fmt.Sprintf("%s,%s", managerType, dnsAddress)
+
+	if err := os.WriteFile(fileUncleanShutdownManagerTypeLocation, []byte(managerData), 0644); err != nil { //nolint:gosec
+		return fmt.Errorf("create %s: %w", fileUncleanShutdownManagerTypeLocation, err)
+	}
+	return nil
+}
+
+func removeUncleanShutdownIndicator() error {
+	if err := os.Remove(fileUncleanShutdownResolvConfLocation); err != nil && !errors.Is(err, fs.ErrNotExist) {
+		return fmt.Errorf("remove %s: %w", fileUncleanShutdownResolvConfLocation, err)
+	}
+	if err := os.Remove(fileUncleanShutdownManagerTypeLocation); err != nil && !errors.Is(err, fs.ErrNotExist) {
+		return fmt.Errorf("remove %s: %w", fileUncleanShutdownManagerTypeLocation, err)
+	}
+	return nil
+}

client/internal/dns/unclean_shutdown_windows.go

@@ -0,0 +1,75 @@
+package dns
+
+import (
+	"errors"
+	"fmt"
+	"io/fs"
+	"os"
+	"path/filepath"
+
+	"github.com/sirupsen/logrus"
+)
+
+const (
+	netbirdProgramDataLocation = "Netbird"
+	fileUncleanShutdownFile    = "unclean_shutdown_dns.txt"
+)
+
+func CheckUncleanShutdown(string) error {
+	file := getUncleanShutdownFile()
+
+	if _, err := os.Stat(file); err != nil {
+		if errors.Is(err, fs.ErrNotExist) {
+			// no file -> clean shutdown
+			return nil
+		} else {
+			return fmt.Errorf("state: %w", err)
+		}
+	}
+
+	logrus.Warnf("detected unclean shutdown, file %s exists. Restoring unclean shutdown dns settings.", file)
+
+	guid, err := os.ReadFile(file)
+	if err != nil {
+		return fmt.Errorf("read %s: %w", file, err)
+	}
+
+	manager, err := newHostManagerWithGuid(string(guid))
+	if err != nil {
+		return fmt.Errorf("create host manager: %w", err)
+	}
+
+	if err := manager.restoreUncleanShutdownDNS(nil); err != nil {
+		return fmt.Errorf("restore unclean shutdown backup: %w", err)
+	}
+
+	return nil
+}
+
+func createUncleanShutdownIndicator(guid string) error {
+	file := getUncleanShutdownFile()
+
+	dir := filepath.Dir(file)
+	if err := os.MkdirAll(dir, os.FileMode(0755)); err != nil {
+		return fmt.Errorf("create dir %s: %w", dir, err)
+	}
+
+	if err := os.WriteFile(file, []byte(guid), 0600); err != nil {
+		return fmt.Errorf("create %s: %w", file, err)
+	}
+
+	return nil
+}
+
+func removeUncleanShutdownIndicator() error {
+	file := getUncleanShutdownFile()
+
+	if err := os.Remove(file); err != nil && !errors.Is(err, fs.ErrNotExist) {
+		return fmt.Errorf("remove %s: %w", file, err)
+	}
+	return nil
+}
+
+func getUncleanShutdownFile() string {
+	return filepath.Join(os.Getenv("PROGRAMDATA"), netbirdProgramDataLocation, fileUncleanShutdownFile)
+}

client/internal/dns/upstream.go

@@ -5,14 +5,16 @@ import (
 	"errors"
 	"fmt"
 	"net"
-	"runtime"
 	"sync"
 	"sync/atomic"
 	"time"
 
 	"github.com/cenkalti/backoff/v4"
+	"github.com/hashicorp/go-multierror"
 	"github.com/miekg/dns"
 	log "github.com/sirupsen/logrus"
+
+	"github.com/netbirdio/netbird/client/internal/peer"
 )
 
 const (
@@ -25,8 +27,7 @@ const (
 const testRecord = "."
 
 type upstreamClient interface {
-	exchange(upstream string, r *dns.Msg) (*dns.Msg, time.Duration, error)
-	exchangeContext(ctx context.Context, upstream string, r *dns.Msg) (*dns.Msg, time.Duration, error)
+	exchange(ctx context.Context, upstream string, r *dns.Msg) (*dns.Msg, time.Duration, error)
 }
 
 type UpstreamResolver interface {
@@ -46,12 +47,13 @@ type upstreamResolverBase struct {
 	reactivatePeriod time.Duration
 	upstreamTimeout  time.Duration
 
-	deactivate func()
-	reactivate func()
+	deactivate     func(error)
+	reactivate     func()
+	statusRecorder *peer.Status
 }
 
-func newUpstreamResolverBase(parentCTX context.Context) *upstreamResolverBase {
-	ctx, cancel := context.WithCancel(parentCTX)
+func newUpstreamResolverBase(ctx context.Context, statusRecorder *peer.Status) *upstreamResolverBase {
+	ctx, cancel := context.WithCancel(ctx)
 
 	return &upstreamResolverBase{
 		ctx:              ctx,
@@ -59,6 +61,7 @@ func newUpstreamResolverBase(parentCTX context.Context) *upstreamResolverBase {
 		upstreamTimeout:  upstreamTimeout,
 		reactivatePeriod: reactivatePeriod,
 		failsTillDeact:   failsTillDeact,
+		statusRecorder:   statusRecorder,
 	}
 }
 
@@ -69,7 +72,10 @@ func (u *upstreamResolverBase) stop() {
 
 // ServeDNS handles a DNS request
 func (u *upstreamResolverBase) ServeDNS(w dns.ResponseWriter, r *dns.Msg) {
-	defer u.checkUpstreamFails()
+	var err error
+	defer func() {
+		u.checkUpstreamFails(err)
+	}()
 
 	log.WithField("question", r.Question[0]).Trace("received an upstream question")
 
@@ -80,8 +86,14 @@ func (u *upstreamResolverBase) ServeDNS(w dns.ResponseWriter, r *dns.Msg) {
 	}
 
 	for _, upstream := range u.upstreamServers {
+		var rm *dns.Msg
+		var t time.Duration
 
-		rm, t, err := u.upstreamClient.exchange(upstream, r)
+		func() {
+			ctx, cancel := context.WithTimeout(u.ctx, u.upstreamTimeout)
+			defer cancel()
+			rm, t, err = u.upstreamClient.exchange(ctx, upstream, r)
+		}()
 
 		if err != nil {
 			if errors.Is(err, context.DeadlineExceeded) || isTimeout(err) {
@@ -126,7 +138,7 @@ func (u *upstreamResolverBase) ServeDNS(w dns.ResponseWriter, r *dns.Msg) {
 // If fails count is greater that failsTillDeact, upstream resolving
 // will be disabled for reactivatePeriod, after that time period fails counter
 // will be reset and upstream will be reactivated.
-func (u *upstreamResolverBase) checkUpstreamFails() {
+func (u *upstreamResolverBase) checkUpstreamFails(err error) {
 	u.mutex.Lock()
 	defer u.mutex.Unlock()
 
@@ -140,7 +152,7 @@ func (u *upstreamResolverBase) checkUpstreamFails() {
 	default:
 	}
 
-	u.disable()
+	u.disable(err)
 }
 
 // probeAvailability tests all upstream servers simultaneously and
@@ -159,13 +171,16 @@ func (u *upstreamResolverBase) probeAvailability() {
 	var mu sync.Mutex
 	var wg sync.WaitGroup
 
+	var errors *multierror.Error
 	for _, upstream := range u.upstreamServers {
 		upstream := upstream
 
 		wg.Add(1)
 		go func() {
 			defer wg.Done()
-			if err := u.testNameserver(upstream); err != nil {
+			err := u.testNameserver(upstream)
+			if err != nil {
+				errors = multierror.Append(errors, err)
 				log.Warnf("probing upstream nameserver %s: %s", upstream, err)
 				return
 			}
@@ -180,7 +195,7 @@ func (u *upstreamResolverBase) probeAvailability() {
 
 	// didn't find a working upstream server, let's disable and try later
 	if !success {
-		u.disable()
+		u.disable(errors.ErrorOrNil())
 	}
 }
 
@@ -213,7 +228,7 @@ func (u *upstreamResolverBase) waitUntilResponse() {
 		}
 
 		log.Tracef("checking connectivity with upstreams %s failed. Retrying in %s", u.upstreamServers, exponentialBackOff.NextBackOff())
-		return fmt.Errorf("got an error from upstream check call")
+		return fmt.Errorf("upstream check call error")
 	}
 
 	err := backoff.Retry(operation, exponentialBackOff)
@@ -239,18 +254,15 @@ func isTimeout(err error) bool {
 	return false
 }
 
-func (u *upstreamResolverBase) disable() {
+func (u *upstreamResolverBase) disable(err error) {
 	if u.disabled {
 		return
 	}
 
-	// todo test the deactivation logic, it seems to affect the client
-	if runtime.GOOS != "ios" {
-		log.Warnf("upstream resolving is Disabled for %v", reactivatePeriod)
-		u.deactivate()
-		u.disabled = true
-		go u.waitUntilResponse()
-	}
+	log.Warnf("Upstream resolving is Disabled for %v", reactivatePeriod)
+	u.deactivate(err)
+	u.disabled = true
+	go u.waitUntilResponse()
 }
 
 func (u *upstreamResolverBase) testNameserver(server string) error {
@@ -259,6 +271,6 @@ func (u *upstreamResolverBase) testNameserver(server string) error {
 
 	r := new(dns.Msg).SetQuestion(testRecord, dns.TypeSOA)
 
-	_, _, err := u.upstreamClient.exchangeContext(ctx, server, r)
+	_, _, err := u.upstreamClient.exchange(ctx, server, r)
 	return err
 }

client/internal/dns/upstream_android.go

@@ -0,0 +1,84 @@
+package dns
+
+import (
+	"context"
+	"net"
+	"syscall"
+	"time"
+
+	"github.com/miekg/dns"
+
+	"github.com/netbirdio/netbird/client/internal/peer"
+	nbnet "github.com/netbirdio/netbird/util/net"
+)
+
+type upstreamResolver struct {
+	*upstreamResolverBase
+	hostsDNSHolder *hostsDNSHolder
+}
+
+// newUpstreamResolver in Android we need to distinguish the DNS servers to available through VPN or outside of VPN
+// In case if the assigned DNS address is available only in the protected network then the resolver will time out at the
+// first time, and we need to wait for a while to start to use again the proper DNS resolver.
+func newUpstreamResolver(
+	ctx context.Context,
+	_ string,
+	_ net.IP,
+	_ *net.IPNet,
+	statusRecorder *peer.Status,
+	hostsDNSHolder *hostsDNSHolder,
+) (*upstreamResolver, error) {
+	upstreamResolverBase := newUpstreamResolverBase(ctx, statusRecorder)
+	c := &upstreamResolver{
+		upstreamResolverBase: upstreamResolverBase,
+		hostsDNSHolder:       hostsDNSHolder,
+	}
+	upstreamResolverBase.upstreamClient = c
+	return c, nil
+}
+
+// exchange in case of Android if the upstream is a local resolver then we do not need to mark the socket as protected.
+// In other case the DNS resolvation goes through the VPN, so we need to force to use the
+func (u *upstreamResolver) exchange(ctx context.Context, upstream string, r *dns.Msg) (rm *dns.Msg, t time.Duration, err error) {
+	if u.isLocalResolver(upstream) {
+		return u.exchangeWithoutVPN(ctx, upstream, r)
+	} else {
+		return u.exchangeWithinVPN(ctx, upstream, r)
+	}
+}
+
+func (u *upstreamResolver) exchangeWithinVPN(ctx context.Context, upstream string, r *dns.Msg) (rm *dns.Msg, t time.Duration, err error) {
+	upstreamExchangeClient := &dns.Client{}
+	return upstreamExchangeClient.ExchangeContext(ctx, r, upstream)
+}
+
+// exchangeWithoutVPN protect the UDP socket by Android SDK to avoid to goes through the VPN
+func (u *upstreamResolver) exchangeWithoutVPN(ctx context.Context, upstream string, r *dns.Msg) (rm *dns.Msg, t time.Duration, err error) {
+	timeout := upstreamTimeout
+	if deadline, ok := ctx.Deadline(); ok {
+		timeout = time.Until(deadline)
+	}
+	dialTimeout := timeout
+
+	nbDialer := nbnet.NewDialer()
+
+	dialer := &net.Dialer{
+		Control: func(network, address string, c syscall.RawConn) error {
+			return nbDialer.Control(network, address, c)
+		},
+		Timeout: dialTimeout,
+	}
+
+	upstreamExchangeClient := &dns.Client{
+		Dialer: dialer,
+	}
+
+	return upstreamExchangeClient.Exchange(r, upstream)
+}
+
+func (u *upstreamResolver) isLocalResolver(upstream string) bool {
+	if u.hostsDNSHolder.isContain(upstream) {
+		return true
+	}
+	return false
+}

client/internal/dns/upstream_general.go

@@ -0,0 +1,38 @@
+//go:build !android && !ios
+
+package dns
+
+import (
+	"context"
+	"net"
+	"time"
+
+	"github.com/miekg/dns"
+
+	"github.com/netbirdio/netbird/client/internal/peer"
+)
+
+type upstreamResolver struct {
+	*upstreamResolverBase
+}
+
+func newUpstreamResolver(
+	ctx context.Context,
+	_ string,
+	_ net.IP,
+	_ *net.IPNet,
+	statusRecorder *peer.Status,
+	_ *hostsDNSHolder,
+) (*upstreamResolver, error) {
+	upstreamResolverBase := newUpstreamResolverBase(ctx, statusRecorder)
+	nonIOS := &upstreamResolver{
+		upstreamResolverBase: upstreamResolverBase,
+	}
+	upstreamResolverBase.upstreamClient = nonIOS
+	return nonIOS, nil
+}
+
+func (u *upstreamResolver) exchange(ctx context.Context, upstream string, r *dns.Msg) (rm *dns.Msg, t time.Duration, err error) {
+	upstreamExchangeClient := &dns.Client{}
+	return upstreamExchangeClient.ExchangeContext(ctx, r, upstream)
+}

client/internal/dns/upstream_ios.go

@@ -11,6 +11,8 @@ import (
 	"github.com/miekg/dns"
 	log "github.com/sirupsen/logrus"
 	"golang.org/x/sys/unix"
+
+	"github.com/netbirdio/netbird/client/internal/peer"
 )
 
 type upstreamResolverIOS struct {
@@ -20,8 +22,15 @@ type upstreamResolverIOS struct {
 	iIndex int
 }
 
-func newUpstreamResolver(parentCTX context.Context, interfaceName string, ip net.IP, net *net.IPNet) (*upstreamResolverIOS, error) {
-	upstreamResolverBase := newUpstreamResolverBase(parentCTX)
+func newUpstreamResolver(
+	ctx context.Context,
+	interfaceName string,
+	ip net.IP,
+	net *net.IPNet,
+	statusRecorder *peer.Status,
+	_ *hostsDNSHolder,
+) (*upstreamResolverIOS, error) {
+	upstreamResolverBase := newUpstreamResolverBase(ctx, statusRecorder)
 
 	index, err := getInterfaceIndex(interfaceName)
 	if err != nil {
@@ -40,34 +49,38 @@ func newUpstreamResolver(parentCTX context.Context, interfaceName string, ip net
 	return ios, nil
 }
 
-func (u *upstreamResolverIOS) exchange(upstream string, r *dns.Msg) (rm *dns.Msg, t time.Duration, err error) {
-	return u.exchangeContext(context.Background(), upstream, r)
-}
-
-func (u *upstreamResolverIOS) exchangeContext(ctx context.Context, upstream string, r *dns.Msg) (rm *dns.Msg, t time.Duration, err error) {
+func (u *upstreamResolverIOS) exchange(ctx context.Context, upstream string, r *dns.Msg) (rm *dns.Msg, t time.Duration, err error) {
 	client := &dns.Client{}
 	upstreamHost, _, err := net.SplitHostPort(upstream)
 	if err != nil {
 		log.Errorf("error while parsing upstream host: %s", err)
 	}
+
+	timeout := upstreamTimeout
+	if deadline, ok := ctx.Deadline(); ok {
+		timeout = time.Until(deadline)
+	}
+	client.DialTimeout = timeout
+
 	upstreamIP := net.ParseIP(upstreamHost)
 	if u.lNet.Contains(upstreamIP) || net.IP.IsPrivate(upstreamIP) {
 		log.Debugf("using private client to query upstream: %s", upstream)
-		client = u.getClientPrivate()
+		client = u.getClientPrivate(timeout)
 	}
 
-	return client.ExchangeContext(ctx, r, upstream)
+	// Cannot use client.ExchangeContext because it overwrites our Dialer
+	return client.Exchange(r, upstream)
 }
 
 // getClientPrivate returns a new DNS client bound to the local IP address of the Netbird interface
 // This method is needed for iOS
-func (u *upstreamResolverIOS) getClientPrivate() *dns.Client {
+func (u *upstreamResolverIOS) getClientPrivate(dialTimeout time.Duration) *dns.Client {
 	dialer := &net.Dialer{
 		LocalAddr: &net.UDPAddr{
 			IP:   u.lIP,
 			Port: 0, // Let the OS pick a free port
 		},
-		Timeout: upstreamTimeout,
+		Timeout: dialTimeout,
 		Control: func(network, address string, c syscall.RawConn) error {
 			var operr error
 			fn := func(s uintptr) {

client/internal/dns/upstream_nonios.go

@@ -1,36 +0,0 @@
-//go:build !ios
-
-package dns
-
-import (
-	"context"
-	"net"
-	"time"
-
-	"github.com/miekg/dns"
-)
-
-type upstreamResolverNonIOS struct {
-	*upstreamResolverBase
-}
-
-func newUpstreamResolver(parentCTX context.Context, interfaceName string, ip net.IP, net *net.IPNet) (*upstreamResolverNonIOS, error) {
-	upstreamResolverBase := newUpstreamResolverBase(parentCTX)
-	nonIOS := &upstreamResolverNonIOS{
-		upstreamResolverBase: upstreamResolverBase,
-	}
-	upstreamResolverBase.upstreamClient = nonIOS
-	return nonIOS, nil
-}
-
-func (u *upstreamResolverNonIOS) exchange(upstream string, r *dns.Msg) (rm *dns.Msg, t time.Duration, err error) {
-	// default upstream timeout
-	ctx, cancel := context.WithTimeout(u.ctx, u.upstreamTimeout)
-	defer cancel()
-	return u.exchangeContext(ctx, upstream, r)
-}
-
-func (u *upstreamResolverNonIOS) exchangeContext(ctx context.Context, upstream string, r *dns.Msg) (rm *dns.Msg, t time.Duration, err error) {
-	upstreamExchangeClient := &dns.Client{}
-	return upstreamExchangeClient.ExchangeContext(ctx, r, upstream)
-}

client/internal/dns/upstream_test.go

@@ -58,7 +58,7 @@ func TestUpstreamResolver_ServeDNS(t *testing.T) {
 	for _, testCase := range testCases {
 		t.Run(testCase.name, func(t *testing.T) {
 			ctx, cancel := context.WithCancel(context.TODO())
-			resolver, _ := newUpstreamResolver(ctx, "", net.IP{}, &net.IPNet{})
+			resolver, _ := newUpstreamResolver(ctx, "", net.IP{}, &net.IPNet{}, nil, nil)
 			resolver.upstreamServers = testCase.InputServers
 			resolver.upstreamTimeout = testCase.timeout
 			if testCase.cancelCTX {
@@ -105,13 +105,8 @@ type mockUpstreamResolver struct {
 	err error
 }
 
-// Exchange mock implementation of Exchangefrom upstreamResolver
-func (c mockUpstreamResolver) exchange(upstream string, r *dns.Msg) (*dns.Msg, time.Duration, error) {
-	return c.exchangeContext(context.Background(), upstream, r)
-}
-
-// ExchangeContext mock implementation of ExchangeContext from upstreamResolver
-func (c mockUpstreamResolver) exchangeContext(_ context.Context, _ string, _ *dns.Msg) (*dns.Msg, time.Duration, error) {
+// exchange mock implementation of exchange from upstreamResolver
+func (c mockUpstreamResolver) exchange(_ context.Context, _ string, _ *dns.Msg) (*dns.Msg, time.Duration, error) {
 	return c.r, c.rtt, c.err
 }
 
@@ -136,7 +131,7 @@ func TestUpstreamResolver_DeactivationReactivation(t *testing.T) {
 	}
 
 	failed := false
-	resolver.deactivate = func() {
+	resolver.deactivate = func(error) {
 		failed = true
 	}
 

client/internal/ebpf/ebpf/dns_fwd_linux.go

@@ -13,7 +13,7 @@ const (
 )
 
 func (tf *GeneralManager) LoadDNSFwd(ip string, dnsPort int) error {
-	log.Debugf("load ebpf DNS forwarder: address: %s:%d", ip, dnsPort)
+	log.Debugf("load eBPF DNS forwarder, watching addr: %s:53, redirect to port: %d", ip, dnsPort)
 	tf.lock.Lock()
 	defer tf.lock.Unlock()
 

client/internal/ebpf/ebpf/src/readme.md

@@ -1,5 +1,10 @@
-# Debug
+# DNS forwarder
 
+The agent attach the XDP program to the lo device. We can not use fake address in eBPF because the
+traffic does not appear in the eBPF program. The program capture the traffic on wg_ip:53 and 
+overwrite in it the destination port to 5053.
+
+# Debug
 
 The CONFIG_BPF_EVENTS kernel module is required for bpf_printk.
 Apply this code to use bpf_printk

client/internal/engine.go

@@ -2,7 +2,9 @@ package internal
 
 import (
 	"context"
+	"errors"
 	"fmt"
+	"maps"
 	"math/rand"
 	"net"
 	"net/netip"
@@ -21,6 +23,7 @@ import (
 	"github.com/netbirdio/netbird/client/firewall/manager"
 	"github.com/netbirdio/netbird/client/internal/acl"
 	"github.com/netbirdio/netbird/client/internal/dns"
+	"github.com/netbirdio/netbird/client/internal/networkmonitor"
 	"github.com/netbirdio/netbird/client/internal/peer"
 	"github.com/netbirdio/netbird/client/internal/relay"
 	"github.com/netbirdio/netbird/client/internal/rosenpass"
@@ -60,6 +63,9 @@ type EngineConfig struct {
 	// WgPrivateKey is a Wireguard private key of our peer (it MUST never leave the machine)
 	WgPrivateKey wgtypes.Key
 
+	// NetworkMonitor is a flag to enable network monitoring
+	NetworkMonitor bool
+
 	// IFaceBlackList is a list of network interfaces to ignore when discovering connection candidates (ICE related)
 	IFaceBlackList       []string
 	DisableIPv6Discovery bool
@@ -79,7 +85,10 @@ type EngineConfig struct {
 
 	CustomDNSAddress string
 
-	RosenpassEnabled bool
+	RosenpassEnabled    bool
+	RosenpassPermissive bool
+
+	ServerSSHAllowed bool
 }
 
 // Engine is a mechanism responsible for reacting on Signal and Management stream events and managing connections to the remote peers.
@@ -90,6 +99,10 @@ type Engine struct {
 	mgmClient mgm.Client
 	// peerConns is a map that holds all the peers that are known to this peer
 	peerConns map[string]*peer.Conn
+
+	beforePeerHook peer.BeforeAddPeerHookFunc
+	afterPeerHook  peer.AfterRemovePeerHookFunc
+
 	// rpManager is a Rosenpass manager
 	rpManager *rosenpass.Manager
 
@@ -104,9 +117,15 @@ type Engine struct {
 	// TURNs is a list of STUN servers used by ICE
 	TURNs []*stun.URI
 
-	cancel context.CancelFunc
+	// clientRoutes is the most recent list of clientRoutes received from the Management Service
+	clientRoutes   route.HAMap
+	clientRoutesMu sync.RWMutex
 
-	ctx context.Context
+	clientCtx    context.Context
+	clientCancel context.CancelFunc
+
+	ctx    context.Context
+	cancel context.CancelFunc
 
 	wgInterface    *iface.WGIface
 	wgProxyFactory *wgproxy.Factory
@@ -116,6 +135,8 @@ type Engine struct {
 	// networkSerial is the latest CurrentSerial (state ID) of the network sent by the Management service
 	networkSerial uint64
 
+	networkMonitor *networkmonitor.NetworkMonitor
+
 	sshServerFunc func(hostKeyPEM []byte, addr string) (nbssh.Server, error)
 	sshServer     nbssh.Server
 
@@ -131,6 +152,8 @@ type Engine struct {
 	signalProbe *Probe
 	relayProbe  *Probe
 	wgProbe     *Probe
+
+	wgConnWorker sync.WaitGroup
 }
 
 // Peer is an instance of the Connection Peer
@@ -141,8 +164,8 @@ type Peer struct {
 
 // NewEngine creates a new Connection Engine
 func NewEngine(
-	ctx context.Context,
-	cancel context.CancelFunc,
+	clientCtx context.Context,
+	clientCancel context.CancelFunc,
 	signalClient signal.Client,
 	mgmClient mgm.Client,
 	config *EngineConfig,
@@ -150,8 +173,8 @@ func NewEngine(
 	statusRecorder *peer.Status,
 ) *Engine {
 	return NewEngineWithProbes(
-		ctx,
-		cancel,
+		clientCtx,
+		clientCancel,
 		signalClient,
 		mgmClient,
 		config,
@@ -166,8 +189,8 @@ func NewEngine(
 
 // NewEngineWithProbes creates a new Connection Engine with probes attached
 func NewEngineWithProbes(
-	ctx context.Context,
-	cancel context.CancelFunc,
+	clientCtx context.Context,
+	clientCancel context.CancelFunc,
 	signalClient signal.Client,
 	mgmClient mgm.Client,
 	config *EngineConfig,
@@ -178,9 +201,10 @@ func NewEngineWithProbes(
 	relayProbe *Probe,
 	wgProbe *Probe,
 ) *Engine {
+
 	return &Engine{
-		ctx:            ctx,
-		cancel:         cancel,
+		clientCtx:      clientCtx,
+		clientCancel:   clientCancel,
 		signal:         signalClient,
 		mgmClient:      mgmClient,
 		peerConns:      make(map[string]*peer.Conn),
@@ -192,7 +216,6 @@ func NewEngineWithProbes(
 		networkSerial:  0,
 		sshServerFunc:  nbssh.DefaultSSHServer,
 		statusRecorder: statusRecorder,
-		wgProxyFactory: wgproxy.NewFactory(config.WgPort),
 		mgmProbe:       mgmProbe,
 		signalProbe:    signalProbe,
 		relayProbe:     relayProbe,
@@ -204,16 +227,31 @@ func (e *Engine) Stop() error {
 	e.syncMsgMux.Lock()
 	defer e.syncMsgMux.Unlock()
 
+	if e.cancel != nil {
+		e.cancel()
+	}
+
+	// stopping network monitor first to avoid starting the engine again
+	if e.networkMonitor != nil {
+		e.networkMonitor.Stop()
+	}
+	log.Info("Network monitor: stopped")
+
 	err := e.removeAllPeers()
 	if err != nil {
 		return err
 	}
 
+	e.clientRoutesMu.Lock()
+	e.clientRoutes = nil
+	e.clientRoutesMu.Unlock()
+
 	// very ugly but we want to remove peers from the WireGuard interface first before removing interface.
-	// Removing peers happens in the conn.CLose() asynchronously
+	// Removing peers happens in the conn.Close() asynchronously
 	time.Sleep(500 * time.Millisecond)
 
 	e.close()
+	e.wgConnWorker.Wait()
 	log.Infof("stopped Netbird Engine")
 	return nil
 }
@@ -225,40 +263,60 @@ func (e *Engine) Start() error {
 	e.syncMsgMux.Lock()
 	defer e.syncMsgMux.Unlock()
 
+	if e.cancel != nil {
+		e.cancel()
+	}
+	e.ctx, e.cancel = context.WithCancel(e.clientCtx)
+
+	e.wgProxyFactory = wgproxy.NewFactory(e.ctx, e.config.WgPort)
+
 	wgIface, err := e.newWgIface()
 	if err != nil {
-		log.Errorf("failed creating wireguard interface instance %s: [%s]", e.config.WgIfaceName, err.Error())
-		return err
+		log.Errorf("failed creating wireguard interface instance %s: [%s]", e.config.WgIfaceName, err)
+		return fmt.Errorf("new wg interface: %w", err)
 	}
 	e.wgInterface = wgIface
 
 	if e.config.RosenpassEnabled {
 		log.Infof("rosenpass is enabled")
+		if e.config.RosenpassPermissive {
+			log.Infof("running rosenpass in permissive mode")
+		} else {
+			log.Infof("running rosenpass in strict mode")
+		}
 		e.rpManager, err = rosenpass.NewManager(e.config.PreSharedKey, e.config.WgIfaceName)
 		if err != nil {
-			return err
+			return fmt.Errorf("create rosenpass manager: %w", err)
 		}
 		err := e.rpManager.Run()
 		if err != nil {
-			return err
+			return fmt.Errorf("run rosenpass manager: %w", err)
 		}
 	}
 
 	initialRoutes, dnsServer, err := e.newDnsServer()
 	if err != nil {
 		e.close()
-		return err
+		return fmt.Errorf("create dns server: %w", err)
 	}
 	e.dnsServer = dnsServer
 
 	e.routeManager = routemanager.NewManager(e.ctx, e.config.WgPrivateKey.PublicKey().String(), e.wgInterface, e.statusRecorder, initialRoutes)
+	beforePeerHook, afterPeerHook, err := e.routeManager.Init()
+	if err != nil {
+		log.Errorf("Failed to initialize route manager: %s", err)
+	} else {
+		e.beforePeerHook = beforePeerHook
+		e.afterPeerHook = afterPeerHook
+	}
+
 	e.routeManager.SetRouteChangeListener(e.mobileDep.NetworkChangeListener)
 
 	err = e.wgInterfaceCreate()
 	if err != nil {
 		log.Errorf("failed creating tunnel interface %s: [%s]", e.config.WgIfaceName, err.Error())
 		e.close()
-		return err
+		return fmt.Errorf("create wg interface: %w", err)
 	}
 
 	e.firewall, err = firewall.NewFirewall(e.ctx, e.wgInterface)
@@ -270,7 +328,7 @@ func (e *Engine) Start() error {
 		err = e.routeManager.EnableServerRouter(e.firewall)
 		if err != nil {
 			e.close()
-			return err
+			return fmt.Errorf("enable server router: %w", err)
 		}
 	}
 
@@ -278,7 +336,7 @@ func (e *Engine) Start() error {
 	if err != nil {
 		log.Errorf("failed to pull up wgInterface [%s]: %s", e.wgInterface.Name(), err.Error())
 		e.close()
-		return err
+		return fmt.Errorf("up wg interface: %w", err)
 	}
 
 	if e.firewall != nil {
@@ -288,13 +346,16 @@ func (e *Engine) Start() error {
 	err = e.dnsServer.Initialize()
 	if err != nil {
 		e.close()
-		return err
+		return fmt.Errorf("initialize dns server: %w", err)
 	}
 
 	e.receiveSignalEvents()
 	e.receiveManagementEvents()
 	e.receiveProbeEvents()
 
+	// starting network monitor at the very last to avoid disruptions
+	e.startNetworkMonitor()
+
 	return nil
 }
 
@@ -482,44 +543,52 @@ func isNil(server nbssh.Server) bool {
 }
 
 func (e *Engine) updateSSH(sshConf *mgmProto.SSHConfig) error {
-	if sshConf.GetSshEnabled() {
-		if runtime.GOOS == "windows" {
-			log.Warnf("running SSH server on Windows is not supported")
-			return nil
-		}
-		// start SSH server if it wasn't running
-		if isNil(e.sshServer) {
-			// nil sshServer means it has not yet been started
-			var err error
-			e.sshServer, err = e.sshServerFunc(e.config.SSHKey,
-				fmt.Sprintf("%s:%d", e.wgInterface.Address().IP.String(), nbssh.DefaultSSHPort))
-			if err != nil {
-				return err
+
+	if !e.config.ServerSSHAllowed {
+		log.Warnf("running SSH server is not permitted")
+		return nil
+	} else {
+
+		if sshConf.GetSshEnabled() {
+			if runtime.GOOS == "windows" {
+				log.Warnf("running SSH server on Windows is not supported")
+				return nil
 			}
-			go func() {
-				// blocking
-				err = e.sshServer.Start()
+			// start SSH server if it wasn't running
+			if isNil(e.sshServer) {
+				// nil sshServer means it has not yet been started
+				var err error
+				e.sshServer, err = e.sshServerFunc(e.config.SSHKey,
+					fmt.Sprintf("%s:%d", e.wgInterface.Address().IP.String(), nbssh.DefaultSSHPort))
 				if err != nil {
-					// will throw error when we stop it even if it is a graceful stop
-					log.Debugf("stopped SSH server with error %v", err)
+					return err
 				}
-				e.syncMsgMux.Lock()
-				defer e.syncMsgMux.Unlock()
-				e.sshServer = nil
-				log.Infof("stopped SSH server")
-			}()
-		} else {
-			log.Debugf("SSH server is already running")
+				go func() {
+					// blocking
+					err = e.sshServer.Start()
+					if err != nil {
+						// will throw error when we stop it even if it is a graceful stop
+						log.Debugf("stopped SSH server with error %v", err)
+					}
+					e.syncMsgMux.Lock()
+					defer e.syncMsgMux.Unlock()
+					e.sshServer = nil
+					log.Infof("stopped SSH server")
+				}()
+			} else {
+				log.Debugf("SSH server is already running")
+			}
+		} else if !isNil(e.sshServer) {
+			// Disable SSH server request, so stop it if it was running
+			err := e.sshServer.Stop()
+			if err != nil {
+				log.Warnf("failed to stop SSH server %v", err)
+			}
+			e.sshServer = nil
 		}
-	} else if !isNil(e.sshServer) {
-		// Disable SSH server request, so stop it if it was running
-		err := e.sshServer.Stop()
-		if err != nil {
-			log.Warnf("failed to stop SSH server %v", err)
-		}
-		e.sshServer = nil
+		return nil
+
 	}
-	return nil
 }
 
 func (e *Engine) updateConfig(conf *mgmProto.PeerConfig) error {
@@ -555,12 +624,12 @@ func (e *Engine) updateConfig(conf *mgmProto.PeerConfig) error {
 // E.g. when a new peer has been registered and we are allowed to connect to it.
 func (e *Engine) receiveManagementEvents() {
 	go func() {
-		err := e.mgmClient.Sync(e.handleSync)
+		err := e.mgmClient.Sync(e.ctx, e.handleSync)
 		if err != nil {
 			// happens if management is unavailable for a long time.
 			// We want to cancel the operation of the whole client
 			_ = CtxGetState(e.ctx).Wrap(ErrResetConnection)
-			e.cancel()
+			e.clientCancel()
 			return
 		}
 		log.Debugf("stopped receiving updates from Management Service")
@@ -667,11 +736,16 @@ func (e *Engine) updateNetworkMap(networkMap *mgmProto.NetworkMap) error {
 	if protoRoutes == nil {
 		protoRoutes = []*mgmProto.Route{}
 	}
-	err := e.routeManager.UpdateRoutes(serial, toRoutes(protoRoutes))
+
+	_, clientRoutes, err := e.routeManager.UpdateRoutes(serial, toRoutes(protoRoutes))
 	if err != nil {
-		log.Errorf("failed to update routes, err: %v", err)
+		log.Errorf("failed to update clientRoutes, err: %v", err)
 	}
 
+	e.clientRoutesMu.Lock()
+	e.clientRoutes = clientRoutes
+	e.clientRoutesMu.Unlock()
+
 	protoDNSConfig := networkMap.GetDNSConfig()
 	if protoDNSConfig == nil {
 		protoDNSConfig = &mgmProto.DNSConfig{}
@@ -682,15 +756,16 @@ func (e *Engine) updateNetworkMap(networkMap *mgmProto.NetworkMap) error {
 		log.Errorf("failed to update dns server, err: %v", err)
 	}
 
-	// Test received (upstream) servers for availability right away instead of upon usage.
-	// If no server of a server group responds this will disable the respective handler and retry later.
-	e.dnsServer.ProbeAvailability()
-
 	if e.acl != nil {
 		e.acl.ApplyFiltering(networkMap)
 	}
+
 	e.networkSerial = serial
 
+	// Test received (upstream) servers for availability right away instead of upon usage.
+	// If no server of a server group responds this will disable the respective handler and retry later.
+	e.dnsServer.ProbeAvailability()
+
 	return nil
 }
 
@@ -699,9 +774,9 @@ func toRoutes(protoRoutes []*mgmProto.Route) []*route.Route {
 	for _, protoRoute := range protoRoutes {
 		_, prefix, _ := route.ParseNetwork(protoRoute.Network)
 		convertedRoute := &route.Route{
-			ID:          protoRoute.ID,
+			ID:          route.ID(protoRoute.ID),
 			Network:     prefix,
-			NetID:       protoRoute.NetID,
+			NetID:       route.NetID(protoRoute.NetID),
 			NetworkType: route.NetworkType(protoRoute.NetworkType),
 			Peer:        protoRoute.Peer,
 			Metric:      int(protoRoute.Metric),
@@ -765,6 +840,7 @@ func (e *Engine) updateOfflinePeers(offlinePeers []*mgmProto.RemotePeerConfig) {
 			FQDN:             offlinePeer.GetFqdn(),
 			ConnStatus:       peer.StatusDisconnected,
 			ConnStatusUpdate: time.Now(),
+			Mux:              new(sync.RWMutex),
 		}
 	}
 	e.statusRecorder.ReplaceOfflinePeers(replacement)
@@ -788,27 +864,39 @@ func (e *Engine) addNewPeer(peerConfig *mgmProto.RemotePeerConfig) error {
 	if _, ok := e.peerConns[peerKey]; !ok {
 		conn, err := e.createPeerConn(peerKey, strings.Join(peerIPs, ","))
 		if err != nil {
-			return err
+			return fmt.Errorf("create peer connection: %w", err)
 		}
 		e.peerConns[peerKey] = conn
 
+		if e.beforePeerHook != nil && e.afterPeerHook != nil {
+			conn.AddBeforeAddPeerHook(e.beforePeerHook)
+			conn.AddAfterRemovePeerHook(e.afterPeerHook)
+		}
+
 		err = e.statusRecorder.AddPeer(peerKey, peerConfig.Fqdn)
 		if err != nil {
 			log.Warnf("error adding peer %s to status recorder, got error: %v", peerKey, err)
 		}
 
+		e.wgConnWorker.Add(1)
 		go e.connWorker(conn, peerKey)
 	}
 	return nil
 }
 
 func (e *Engine) connWorker(conn *peer.Conn, peerKey string) {
+	defer e.wgConnWorker.Done()
 	for {
 
 		// randomize starting time a bit
 		min := 500
 		max := 2000
-		time.Sleep(time.Duration(rand.Intn(max-min)+min) * time.Millisecond)
+		duration := time.Duration(rand.Intn(max-min)+min) * time.Millisecond
+		select {
+		case <-e.ctx.Done():
+			return
+		case <-time.After(duration):
+		}
 
 		// if peer has been removed -> give up
 		if !e.peerExists(peerKey) {
@@ -826,11 +914,12 @@ func (e *Engine) connWorker(conn *peer.Conn, peerKey string) {
 		conn.UpdateStunTurn(append(e.STUNs, e.TURNs...))
 		e.syncMsgMux.Unlock()
 
-		err := conn.Open()
+		err := conn.Open(e.ctx)
 		if err != nil {
 			log.Debugf("connection to peer %s failed: %v", peerKey, err)
-			switch err.(type) {
-			case *peer.ConnectionClosedError:
+			var connectionClosedError *peer.ConnectionClosedError
+			switch {
+			case errors.As(err, &connectionClosedError):
 				// conn has been forced to close, so we exit the loop
 				return
 			default:
@@ -860,7 +949,7 @@ func (e *Engine) createPeerConn(pubKey string, allowedIPs string) (*peer.Conn, e
 		PreSharedKey: e.config.PreSharedKey,
 	}
 
-	if e.config.RosenpassEnabled {
+	if e.config.RosenpassEnabled && !e.config.RosenpassPermissive {
 		lk := []byte(e.config.WgPrivateKey.PublicKey().String())
 		rk := []byte(wgConfig.RemoteKey)
 		var keyInput []byte
@@ -894,7 +983,6 @@ func (e *Engine) createPeerConn(pubKey string, allowedIPs string) (*peer.Conn, e
 		WgConfig:             wgConfig,
 		LocalWgPort:          e.config.WgPort,
 		NATExternalIPs:       e.parseNATExternalIPMappings(),
-		UserspaceBind:        e.wgInterface.IsUserspaceBind(),
 		RosenpassPubKey:      e.getRosenpassPubKey(),
 		RosenpassAddr:        e.getRosenpassAddr(),
 	}
@@ -941,7 +1029,7 @@ func (e *Engine) createPeerConn(pubKey string, allowedIPs string) (*peer.Conn, e
 func (e *Engine) receiveSignalEvents() {
 	go func() {
 		// connect to a stream of messages coming from the signal server
-		err := e.signal.Receive(func(msg *sProto.Message) error {
+		err := e.signal.Receive(e.ctx, func(msg *sProto.Message) error {
 			e.syncMsgMux.Lock()
 			defer e.syncMsgMux.Unlock()
 
@@ -957,8 +1045,6 @@ func (e *Engine) receiveSignalEvents() {
 					return err
 				}
 
-				conn.RegisterProtoSupportMeta(msg.Body.GetFeaturesSupported())
-
 				var rosenpassPubKey []byte
 				rosenpassAddr := ""
 				if msg.GetBody().GetRosenpassConfig() != nil {
@@ -981,8 +1067,6 @@ func (e *Engine) receiveSignalEvents() {
 					return err
 				}
 
-				conn.RegisterProtoSupportMeta(msg.GetBody().GetFeaturesSupported())
-
 				var rosenpassPubKey []byte
 				rosenpassAddr := ""
 				if msg.GetBody().GetRosenpassConfig() != nil {
@@ -1005,7 +1089,8 @@ func (e *Engine) receiveSignalEvents() {
 					log.Errorf("failed on parsing remote candidate %s -> %s", candidate, err)
 					return err
 				}
-				conn.OnRemoteCandidate(candidate)
+
+				conn.OnRemoteCandidate(candidate, e.GetClientRoutes())
 			case sProto.Body_MODE:
 			}
 
@@ -1015,7 +1100,7 @@ func (e *Engine) receiveSignalEvents() {
 			// happens if signal is unavailable for a long time.
 			// We want to cancel the operation of the whole client
 			_ = CtxGetState(e.ctx).Wrap(ErrResetConnection)
-			e.cancel()
+			e.clientCancel()
 			return
 		}
 	}()
@@ -1076,8 +1161,20 @@ func (e *Engine) parseNATExternalIPMappings() []string {
 }
 
 func (e *Engine) close() {
-	if err := e.wgProxyFactory.Free(); err != nil {
-		log.Errorf("failed closing ebpf proxy: %s", err)
+	if e.wgProxyFactory != nil {
+		if err := e.wgProxyFactory.Free(); err != nil {
+			log.Errorf("failed closing ebpf proxy: %s", err)
+		}
+	}
+
+	// stop/restore DNS first so dbus and friends don't complain because of a missing interface
+	if e.dnsServer != nil {
+		e.dnsServer.Stop()
+		e.dnsServer = nil
+	}
+
+	if e.routeManager != nil {
+		e.routeManager.Stop()
 	}
 
 	log.Debugf("removing Netbird interface %s", e.config.WgIfaceName)
@@ -1094,14 +1191,6 @@ func (e *Engine) close() {
 		}
 	}
 
-	if e.routeManager != nil {
-		e.routeManager.Stop()
-	}
-
-	if e.dnsServer != nil {
-		e.dnsServer.Stop()
-	}
-
 	if e.firewall != nil {
 		err := e.firewall.Reset()
 		if err != nil {
@@ -1171,14 +1260,21 @@ func (e *Engine) newDnsServer() ([]*route.Route, dns.Server, error) {
 		if err != nil {
 			return nil, nil, err
 		}
-		dnsServer := dns.NewDefaultServerPermanentUpstream(e.ctx, e.wgInterface, e.mobileDep.HostDNSAddresses, *dnsConfig, e.mobileDep.NetworkChangeListener)
+		dnsServer := dns.NewDefaultServerPermanentUpstream(
+			e.ctx,
+			e.wgInterface,
+			e.mobileDep.HostDNSAddresses,
+			*dnsConfig,
+			e.mobileDep.NetworkChangeListener,
+			e.statusRecorder,
+		)
 		go e.mobileDep.DnsReadyListener.OnReady()
 		return routes, dnsServer, nil
 	case "ios":
-		dnsServer := dns.NewDefaultServerIos(e.ctx, e.wgInterface, e.mobileDep.DnsManager)
+		dnsServer := dns.NewDefaultServerIos(e.ctx, e.wgInterface, e.mobileDep.DnsManager, e.statusRecorder)
 		return nil, dnsServer, nil
 	default:
-		dnsServer, err := dns.NewDefaultServer(e.ctx, e.wgInterface, e.config.CustomDNSAddress)
+		dnsServer, err := dns.NewDefaultServer(e.ctx, e.wgInterface, e.config.CustomDNSAddress, e.statusRecorder)
 		if err != nil {
 			return nil, nil, err
 		}
@@ -1186,6 +1282,31 @@ func (e *Engine) newDnsServer() ([]*route.Route, dns.Server, error) {
 	}
 }
 
+// GetClientRoutes returns the current routes from the route map
+func (e *Engine) GetClientRoutes() route.HAMap {
+	e.clientRoutesMu.RLock()
+	defer e.clientRoutesMu.RUnlock()
+
+	return maps.Clone(e.clientRoutes)
+}
+
+// GetClientRoutesWithNetID returns the current routes from the route map, but the keys consist of the network ID only
+func (e *Engine) GetClientRoutesWithNetID() map[route.NetID][]*route.Route {
+	e.clientRoutesMu.RLock()
+	defer e.clientRoutesMu.RUnlock()
+
+	routes := make(map[route.NetID][]*route.Route, len(e.clientRoutes))
+	for id, v := range e.clientRoutes {
+		routes[id.NetID()] = v
+	}
+	return routes
+}
+
+// GetRouteManager returns the route manager
+func (e *Engine) GetRouteManager() routemanager.Manager {
+	return e.routeManager
+}
+
 func findIPFromInterfaceName(ifaceName string) (net.IP, error) {
 	iface, err := net.InterfaceByName(ifaceName)
 	if err != nil {
@@ -1269,7 +1390,7 @@ func (e *Engine) receiveProbeEvents() {
 					log.Debugf("failed to get wg stats for peer %s: %s", key, err)
 				}
 				// wgStats could be zero value, in which case we just reset the stats
-				if err := e.statusRecorder.UpdateWireguardPeerState(key, wgStats); err != nil {
+				if err := e.statusRecorder.UpdateWireGuardPeerState(key, wgStats); err != nil {
 					log.Debugf("failed to update wg stats for peer %s: %s", key, err)
 				}
 			}
@@ -1286,3 +1407,26 @@ func (e *Engine) probeSTUNs() []relay.ProbeResult {
 func (e *Engine) probeTURNs() []relay.ProbeResult {
 	return relay.ProbeAll(e.ctx, relay.ProbeTURN, e.TURNs)
 }
+
+func (e *Engine) startNetworkMonitor() {
+	if !e.config.NetworkMonitor {
+		log.Infof("Network monitor is disabled, not starting")
+		return
+	}
+
+	e.networkMonitor = networkmonitor.New()
+	go func() {
+		err := e.networkMonitor.Start(e.ctx, func() {
+			log.Infof("Network monitor detected network change, restarting engine")
+			if err := e.Stop(); err != nil {
+				log.Errorf("Failed to stop engine: %v", err)
+			}
+			if err := e.Start(); err != nil {
+				log.Errorf("Failed to start engine: %v", err)
+			}
+		})
+		if err != nil && !errors.Is(err, networkmonitor.ErrStopped) {
+			log.Errorf("Network monitor: %v", err)
+		}
+	}()
+}

client/internal/engine_test.go

@@ -21,6 +21,8 @@ import (
 	"google.golang.org/grpc"
 	"google.golang.org/grpc/keepalive"
 
+	"github.com/netbirdio/management-integrations/integrations"
+
 	"github.com/netbirdio/netbird/client/internal/dns"
 	"github.com/netbirdio/netbird/client/internal/peer"
 	"github.com/netbirdio/netbird/client/internal/routemanager"
@@ -70,10 +72,11 @@ func TestEngine_SSH(t *testing.T) {
 	defer cancel()
 
 	engine := NewEngine(ctx, cancel, &signal.MockClient{}, &mgmt.MockClient{}, &EngineConfig{
-		WgIfaceName:  "utun101",
-		WgAddr:       "100.64.0.1/24",
-		WgPrivateKey: key,
-		WgPort:       33100,
+		WgIfaceName:      "utun101",
+		WgAddr:           "100.64.0.1/24",
+		WgPrivateKey:     key,
+		WgPort:           33100,
+		ServerSSHAllowed: true,
 	}, MobileDependency{}, peer.NewRecorder("https://mgm"))
 
 	engine.dnsServer = &dns.MockServer{
@@ -226,6 +229,7 @@ func TestEngine_UpdateNetworkMap(t *testing.T) {
 		t.Fatal(err)
 	}
 	engine.udpMux = bind.NewUniversalUDPMuxDefault(bind.UniversalUDPMuxParams{UDPConn: conn})
+	engine.ctx = ctx
 
 	type testCase struct {
 		name       string
@@ -389,7 +393,7 @@ func TestEngine_Sync(t *testing.T) {
 	// feed updates to Engine via mocked Management client
 	updates := make(chan *mgmtProto.SyncResponse)
 	defer close(updates)
-	syncFunc := func(msgHandler func(msg *mgmtProto.SyncResponse) error) error {
+	syncFunc := func(ctx context.Context, msgHandler func(msg *mgmtProto.SyncResponse) error) error {
 		for msg := range updates {
 			err := msgHandler(msg)
 			if err != nil {
@@ -405,6 +409,7 @@ func TestEngine_Sync(t *testing.T) {
 		WgPrivateKey: key,
 		WgPort:       33100,
 	}, MobileDependency{}, peer.NewRecorder("https://mgm"))
+	engine.ctx = ctx
 
 	engine.dnsServer = &dns.MockServer{
 		UpdateDNSServerFunc: func(serial uint64, update nbdns.Config) error { return nil },
@@ -563,6 +568,7 @@ func TestEngine_UpdateNetworkMapWithRoutes(t *testing.T) {
 				WgPrivateKey: key,
 				WgPort:       33100,
 			}, MobileDependency{}, peer.NewRecorder("https://mgm"))
+			engine.ctx = ctx
 			newNet, err := stdnet.NewNet()
 			if err != nil {
 				t.Fatal(err)
@@ -575,10 +581,10 @@ func TestEngine_UpdateNetworkMapWithRoutes(t *testing.T) {
 			}{}
 
 			mockRouteManager := &routemanager.MockManager{
-				UpdateRoutesFunc: func(updateSerial uint64, newRoutes []*route.Route) error {
+				UpdateRoutesFunc: func(updateSerial uint64, newRoutes []*route.Route) (map[route.ID]*route.Route, route.HAMap, error) {
 					input.inputSerial = updateSerial
 					input.inputRoutes = newRoutes
-					return testCase.inputErr
+					return nil, nil, testCase.inputErr
 				},
 			}
 
@@ -595,8 +601,8 @@ func TestEngine_UpdateNetworkMapWithRoutes(t *testing.T) {
 			err = engine.updateNetworkMap(testCase.networkMap)
 			assert.NoError(t, err, "shouldn't return error")
 			assert.Equal(t, testCase.expectedSerial, input.inputSerial, "serial should match")
-			assert.Len(t, input.inputRoutes, testCase.expectedLen, "routes len should match")
-			assert.Equal(t, testCase.expectedRoutes, input.inputRoutes, "routes should match")
+			assert.Len(t, input.inputRoutes, testCase.expectedLen, "clientRoutes len should match")
+			assert.Equal(t, testCase.expectedRoutes, input.inputRoutes, "clientRoutes should match")
 		})
 	}
 }
@@ -732,6 +738,8 @@ func TestEngine_UpdateNetworkMapWithDNSUpdate(t *testing.T) {
 				WgPrivateKey: key,
 				WgPort:       33100,
 			}, MobileDependency{}, peer.NewRecorder("https://mgm"))
+			engine.ctx = ctx
+
 			newNet, err := stdnet.NewNet()
 			if err != nil {
 				t.Fatal(err)
@@ -740,8 +748,8 @@ func TestEngine_UpdateNetworkMapWithDNSUpdate(t *testing.T) {
 			assert.NoError(t, err, "shouldn't return error")
 
 			mockRouteManager := &routemanager.MockManager{
-				UpdateRoutesFunc: func(updateSerial uint64, newRoutes []*route.Route) error {
-					return nil
+				UpdateRoutesFunc: func(updateSerial uint64, newRoutes []*route.Route) (map[route.ID]*route.Route, route.HAMap, error) {
+					return nil, nil, nil
 				},
 			}
 
@@ -1000,7 +1008,9 @@ func createEngine(ctx context.Context, cancel context.CancelFunc, setupKey strin
 		WgPort:       wgPort,
 	}
 
-	return NewEngine(ctx, cancel, signalClient, mgmtClient, conf, MobileDependency{}, peer.NewRecorder("https://mgm")), nil
+	e, err := NewEngine(ctx, cancel, signalClient, mgmtClient, conf, MobileDependency{}, peer.NewRecorder("https://mgm")), nil
+	e.ctx = ctx
+	return e, err
 }
 
 func startSignal() (*grpc.Server, string, error) {
@@ -1039,7 +1049,7 @@ func startManagement(dataDir string) (*grpc.Server, string, error) {
 		return nil, "", err
 	}
 	s := grpc.NewServer(grpc.KeepaliveEnforcementPolicy(kaep), grpc.KeepaliveParams(kasp))
-	store, err := server.NewStoreFromJson(config.Datadir, nil)
+	store, _, err := server.NewTestStoreFromJson(config.Datadir)
 	if err != nil {
 		return nil, "", err
 	}
@@ -1049,8 +1059,8 @@ func startManagement(dataDir string) (*grpc.Server, string, error) {
 	if err != nil {
 		return nil, "", err
 	}
-	accountManager, err := server.BuildManager(store, peersUpdateManager, nil, "", "",
-		eventStore, false)
+	ia, _ := integrations.NewIntegratedValidator(eventStore)
+	accountManager, err := server.BuildManager(store, peersUpdateManager, nil, "", "netbird.selfhosted", eventStore, nil, false, ia)
 	if err != nil {
 		return nil, "", err
 	}

client/internal/login.go

@@ -68,7 +68,7 @@ func Login(ctx context.Context, config *Config, setupKey string, jwtToken string
 	}
 
 	serverKey, err := doMgmLogin(ctx, mgmClient, pubSSHKey)
-	if isRegistrationNeeded(err) {
+	if serverKey != nil && isRegistrationNeeded(err) {
 		log.Debugf("peer registration required")
 		_, err = registerPeer(ctx, *serverKey, mgmClient, setupKey, jwtToken, pubSSHKey)
 		return err

client/internal/networkmonitor/monitor.go

@@ -0,0 +1,21 @@
+package networkmonitor
+
+import (
+	"context"
+	"errors"
+	"sync"
+)
+
+var ErrStopped = errors.New("monitor has been stopped")
+
+// NetworkMonitor watches for changes in network configuration.
+type NetworkMonitor struct {
+	cancel context.CancelFunc
+	wg     sync.WaitGroup
+	mu     sync.Mutex
+}
+
+// New creates a new network monitor.
+func New() *NetworkMonitor {
+	return &NetworkMonitor{}
+}

client/internal/networkmonitor/monitor_bsd.go

@@ -0,0 +1,133 @@
+//go:build (darwin && !ios) || dragonfly || freebsd || netbsd || openbsd
+
+package networkmonitor
+
+import (
+	"context"
+	"fmt"
+	"net"
+	"net/netip"
+	"syscall"
+	"unsafe"
+
+	log "github.com/sirupsen/logrus"
+	"golang.org/x/net/route"
+	"golang.org/x/sys/unix"
+
+	"github.com/netbirdio/netbird/client/internal/routemanager"
+)
+
+func checkChange(ctx context.Context, nexthopv4 netip.Addr, intfv4 *net.Interface, nexthopv6 netip.Addr, intfv6 *net.Interface, callback func()) error {
+	fd, err := unix.Socket(syscall.AF_ROUTE, syscall.SOCK_RAW, syscall.AF_UNSPEC)
+	if err != nil {
+		return fmt.Errorf("failed to open routing socket: %v", err)
+	}
+	defer func() {
+		if err := unix.Close(fd); err != nil {
+			log.Errorf("Network monitor: failed to close routing socket: %v", err)
+		}
+	}()
+
+	for {
+		select {
+		case <-ctx.Done():
+			return ErrStopped
+		default:
+			buf := make([]byte, 2048)
+			n, err := unix.Read(fd, buf)
+			if err != nil {
+				log.Errorf("Network monitor: failed to read from routing socket: %v", err)
+				continue
+			}
+			if n < unix.SizeofRtMsghdr {
+				log.Errorf("Network monitor: read from routing socket returned less than expected: %d bytes", n)
+				continue
+			}
+
+			msg := (*unix.RtMsghdr)(unsafe.Pointer(&buf[0]))
+
+			switch msg.Type {
+
+			// handle interface state changes
+			case unix.RTM_IFINFO:
+				ifinfo, err := parseInterfaceMessage(buf[:n])
+				if err != nil {
+					log.Errorf("Network monitor: error parsing interface message: %v", err)
+					continue
+				}
+				if msg.Flags&unix.IFF_UP != 0 {
+					continue
+				}
+				if (intfv4 == nil || ifinfo.Index != intfv4.Index) && (intfv6 == nil || ifinfo.Index != intfv6.Index) {
+					continue
+				}
+
+				log.Infof("Network monitor: monitored interface (%s) is down.", ifinfo.Name)
+				go callback()
+
+			// handle route changes
+			case unix.RTM_ADD, syscall.RTM_DELETE:
+				route, err := parseRouteMessage(buf[:n])
+				if err != nil {
+					log.Errorf("Network monitor: error parsing routing message: %v", err)
+					continue
+				}
+
+				if !route.Dst.Addr().IsUnspecified() {
+					continue
+				}
+
+				intf := "<nil>"
+				if route.Interface != nil {
+					intf = route.Interface.Name
+				}
+				switch msg.Type {
+				case unix.RTM_ADD:
+					log.Infof("Network monitor: default route changed: via %s, interface %s", route.Gw, intf)
+					go callback()
+				case unix.RTM_DELETE:
+					if intfv4 != nil && route.Gw.Compare(nexthopv4) == 0 || intfv6 != nil && route.Gw.Compare(nexthopv6) == 0 {
+						log.Infof("Network monitor: default route removed: via %s, interface %s", route.Gw, intf)
+						go callback()
+					}
+				}
+			}
+		}
+	}
+}
+
+func parseInterfaceMessage(buf []byte) (*route.InterfaceMessage, error) {
+	msgs, err := route.ParseRIB(route.RIBTypeInterface, buf)
+	if err != nil {
+		return nil, fmt.Errorf("parse RIB: %v", err)
+	}
+
+	if len(msgs) != 1 {
+		return nil, fmt.Errorf("unexpected RIB message msgs: %v", msgs)
+	}
+
+	msg, ok := msgs[0].(*route.InterfaceMessage)
+	if !ok {
+		return nil, fmt.Errorf("unexpected RIB message type: %T", msgs[0])
+	}
+
+	return msg, nil
+}
+
+func parseRouteMessage(buf []byte) (*routemanager.Route, error) {
+	msgs, err := route.ParseRIB(route.RIBTypeRoute, buf)
+	if err != nil {
+		return nil, fmt.Errorf("parse RIB: %v", err)
+	}
+
+	if len(msgs) != 1 {
+		return nil, fmt.Errorf("unexpected RIB message msgs: %v", msgs)
+	}
+
+	msg, ok := msgs[0].(*route.RouteMessage)
+	if !ok {
+		return nil, fmt.Errorf("unexpected RIB message type: %T", msgs[0])
+	}
+
+	return routemanager.MsgToRoute(msg)
+}

client/internal/networkmonitor/monitor_generic.go

@@ -0,0 +1,84 @@
+//go:build !ios && !android
+
+package networkmonitor
+
+import (
+	"context"
+	"errors"
+	"fmt"
+	"net"
+	"net/netip"
+	"runtime/debug"
+
+	"github.com/cenkalti/backoff/v4"
+	log "github.com/sirupsen/logrus"
+
+	"github.com/netbirdio/netbird/client/internal/routemanager"
+)
+
+// Start begins monitoring network changes. When a change is detected, it calls the callback asynchronously and returns.
+func (nw *NetworkMonitor) Start(ctx context.Context, callback func()) (err error) {
+	if ctx.Err() != nil {
+		return ctx.Err()
+	}
+
+	nw.mu.Lock()
+	ctx, nw.cancel = context.WithCancel(ctx)
+	nw.mu.Unlock()
+
+	nw.wg.Add(1)
+	defer nw.wg.Done()
+
+	var nexthop4, nexthop6 netip.Addr
+	var intf4, intf6 *net.Interface
+
+	operation := func() error {
+		var errv4, errv6 error
+		nexthop4, intf4, errv4 = routemanager.GetNextHop(netip.IPv4Unspecified())
+		nexthop6, intf6, errv6 = routemanager.GetNextHop(netip.IPv6Unspecified())
+
+		if errv4 != nil && errv6 != nil {
+			return errors.New("failed to get default next hops")
+		}
+
+		if errv4 == nil {
+			log.Debugf("Network monitor: IPv4 default route: %s, interface: %s", nexthop4, intf4.Name)
+		}
+		if errv6 == nil {
+			log.Debugf("Network monitor: IPv6 default route: %s, interface: %s", nexthop6, intf6.Name)
+		}
+
+		// continue if either route was found
+		return nil
+	}
+
+	expBackOff := backoff.WithContext(backoff.NewExponentialBackOff(), ctx)
+
+	if err := backoff.Retry(operation, expBackOff); err != nil {
+		return fmt.Errorf("failed to get default next hops: %w", err)
+	}
+
+	// recover in case sys ops panic
+	defer func() {
+		if r := recover(); r != nil {
+			err = fmt.Errorf("panic occurred: %v, stack trace: %s", r, string(debug.Stack()))
+		}
+	}()
+
+	if err := checkChange(ctx, nexthop4, intf4, nexthop6, intf6, callback); err != nil {
+		return fmt.Errorf("check change: %w", err)
+	}
+
+	return nil
+}
+
+// Stop stops the network monitor.
+func (nw *NetworkMonitor) Stop() {
+	nw.mu.Lock()
+	defer nw.mu.Unlock()
+
+	if nw.cancel != nil {
+		nw.cancel()
+		nw.wg.Wait()
+	}
+}

client/internal/networkmonitor/monitor_linux.go

@@ -0,0 +1,81 @@
+//go:build !android
+
+package networkmonitor
+
+import (
+	"context"
+	"errors"
+	"fmt"
+	"net"
+	"net/netip"
+	"syscall"
+
+	log "github.com/sirupsen/logrus"
+	"github.com/vishvananda/netlink"
+)
+
+func checkChange(ctx context.Context, nexthopv4 netip.Addr, intfv4 *net.Interface, nexthop6 netip.Addr, intfv6 *net.Interface, callback func()) error {
+	if intfv4 == nil && intfv6 == nil {
+		return errors.New("no interfaces available")
+	}
+
+	linkChan := make(chan netlink.LinkUpdate)
+	done := make(chan struct{})
+	defer close(done)
+
+	if err := netlink.LinkSubscribe(linkChan, done); err != nil {
+		return fmt.Errorf("subscribe to link updates: %v", err)
+	}
+
+	routeChan := make(chan netlink.RouteUpdate)
+	if err := netlink.RouteSubscribe(routeChan, done); err != nil {
+		return fmt.Errorf("subscribe to route updates: %v", err)
+	}
+
+	log.Info("Network monitor: started")
+	for {
+		select {
+		case <-ctx.Done():
+			return ErrStopped
+
+		// handle interface state changes
+		case update := <-linkChan:
+			if (intfv4 == nil || update.Index != int32(intfv4.Index)) && (intfv6 == nil || update.Index != int32(intfv6.Index)) {
+				continue
+			}
+
+			switch update.Header.Type {
+			case syscall.RTM_DELLINK:
+				log.Infof("Network monitor: monitored interface (%s) is gone", update.Link.Attrs().Name)
+				go callback()
+				return nil
+			case syscall.RTM_NEWLINK:
+				if (update.IfInfomsg.Flags&syscall.IFF_RUNNING) == 0 && update.Link.Attrs().OperState == netlink.OperDown {
+					log.Infof("Network monitor: monitored interface (%s) is down.", update.Link.Attrs().Name)
+					go callback()
+					return nil
+				}
+			}
+
+		// handle route changes
+		case route := <-routeChan:
+			// default route and main table
+			if route.Dst != nil || route.Table != syscall.RT_TABLE_MAIN {
+				continue
+			}
+			switch route.Type {
+			// triggered on added/replaced routes
+			case syscall.RTM_NEWROUTE:
+				log.Infof("Network monitor: default route changed: via %s, interface %d", route.Gw, route.LinkIndex)
+				go callback()
+				return nil
+			case syscall.RTM_DELROUTE:
+				if intfv4 != nil && route.Gw.Equal(nexthopv4.AsSlice()) || intfv6 != nil && route.Gw.Equal(nexthop6.AsSlice()) {
+					log.Infof("Network monitor: default route removed: via %s, interface %d", route.Gw, route.LinkIndex)
+					go callback()
+					return nil
+				}
+			}
+		}
+	}
+}

client/internal/networkmonitor/monitor_mobile.go

@@ -0,0 +1,12 @@
+//go:build ios || android
+
+package networkmonitor
+
+import "context"
+
+func (nw *NetworkMonitor) Start(context.Context, func()) error {
+	return nil
+}
+
+func (nw *NetworkMonitor) Stop() {
+}

client/internal/networkmonitor/monitor_windows.go

@@ -0,0 +1,215 @@
+package networkmonitor
+
+import (
+	"context"
+	"fmt"
+	"net"
+	"net/netip"
+	"time"
+
+	log "github.com/sirupsen/logrus"
+
+	"github.com/netbirdio/netbird/client/internal/routemanager"
+)
+
+const (
+	unreachable = 0
+	incomplete  = 1
+	probe       = 2
+	delay       = 3
+	stale       = 4
+	reachable   = 5
+	permanent   = 6
+	tbd         = 7
+)
+
+const interval = 10 * time.Second
+
+func checkChange(ctx context.Context, nexthopv4 netip.Addr, intfv4 *net.Interface, nexthopv6 netip.Addr, intfv6 *net.Interface, callback func()) error {
+	var neighborv4, neighborv6 *routemanager.Neighbor
+	{
+		initialNeighbors, err := getNeighbors()
+		if err != nil {
+			return fmt.Errorf("get neighbors: %w", err)
+		}
+
+		if n, ok := initialNeighbors[nexthopv4]; ok {
+			neighborv4 = &n
+		}
+		if n, ok := initialNeighbors[nexthopv6]; ok {
+			neighborv6 = &n
+		}
+	}
+	log.Debugf("Network monitor: initial IPv4 neighbor: %v, IPv6 neighbor: %v", neighborv4, neighborv6)
+
+	ticker := time.NewTicker(interval)
+	defer ticker.Stop()
+
+	for {
+		select {
+		case <-ctx.Done():
+			return ErrStopped
+		case <-ticker.C:
+			if changed(nexthopv4, intfv4, neighborv4, nexthopv6, intfv6, neighborv6) {
+				go callback()
+				return nil
+			}
+		}
+	}
+}
+
+func changed(
+	nexthopv4 netip.Addr,
+	intfv4 *net.Interface,
+	neighborv4 *routemanager.Neighbor,
+	nexthopv6 netip.Addr,
+	intfv6 *net.Interface,
+	neighborv6 *routemanager.Neighbor,
+) bool {
+	neighbors, err := getNeighbors()
+	if err != nil {
+		log.Errorf("network monitor: error fetching current neighbors: %v", err)
+		return false
+	}
+	if neighborChanged(nexthopv4, neighborv4, neighbors) || neighborChanged(nexthopv6, neighborv6, neighbors) {
+		return true
+	}
+
+	routes, err := getRoutes()
+	if err != nil {
+		log.Errorf("network monitor: error fetching current routes: %v", err)
+		return false
+	}
+
+	if routeChanged(nexthopv4, intfv4, routes) || routeChanged(nexthopv6, intfv6, routes) {
+		return true
+	}
+
+	return false
+}
+
+// routeChanged checks if the default routes still point to our nexthop/interface
+func routeChanged(nexthop netip.Addr, intf *net.Interface, routes map[netip.Prefix]routemanager.Route) bool {
+	if !nexthop.IsValid() {
+		return false
+	}
+
+	var unspec netip.Prefix
+	if nexthop.Is6() {
+		unspec = netip.PrefixFrom(netip.IPv6Unspecified(), 0)
+	} else {
+		unspec = netip.PrefixFrom(netip.IPv4Unspecified(), 0)
+	}
+
+	if r, ok := routes[unspec]; ok {
+		if r.Nexthop != nexthop || compareIntf(r.Interface, intf) != 0 {
+			intf := "<nil>"
+			if r.Interface != nil {
+				intf = r.Interface.Name
+			}
+			log.Infof("network monitor: default route changed: %s via %s (%s)", r.Destination, r.Nexthop, intf)
+			return true
+		}
+	} else {
+		log.Infof("network monitor: default route is gone")
+		return true
+	}
+
+	return false
+
+}
+
+func neighborChanged(nexthop netip.Addr, neighbor *routemanager.Neighbor, neighbors map[netip.Addr]routemanager.Neighbor) bool {
+	if neighbor == nil {
+		return false
+	}
+
+	// TODO: consider non-local nexthops, e.g. on point-to-point interfaces
+	if n, ok := neighbors[nexthop]; ok {
+		if n.State != reachable && n.State != permanent {
+			log.Infof("network monitor: neighbor %s (%s) is not reachable: %s", neighbor.IPAddress, neighbor.LinkLayerAddress, stateFromInt(n.State))
+			return true
+		} else if n.InterfaceIndex != neighbor.InterfaceIndex {
+			log.Infof(
+				"network monitor: neighbor %s (%s) changed interface from '%s' (%d) to '%s' (%d): %s",
+				neighbor.IPAddress,
+				neighbor.LinkLayerAddress,
+				neighbor.InterfaceAlias,
+				neighbor.InterfaceIndex,
+				n.InterfaceAlias,
+				n.InterfaceIndex,
+				stateFromInt(n.State),
+			)
+			return true
+		}
+	} else {
+		log.Infof("network monitor: neighbor %s (%s) is gone", neighbor.IPAddress, neighbor.LinkLayerAddress)
+		return true
+	}
+
+	return false
+}
+
+func getNeighbors() (map[netip.Addr]routemanager.Neighbor, error) {
+	entries, err := routemanager.GetNeighbors()
+	if err != nil {
+		return nil, fmt.Errorf("get neighbors: %w", err)
+	}
+
+	neighbours := make(map[netip.Addr]routemanager.Neighbor, len(entries))
+	for _, entry := range entries {
+		neighbours[entry.IPAddress] = entry
+	}
+
+	return neighbours, nil
+}
+
+func getRoutes() (map[netip.Prefix]routemanager.Route, error) {
+	entries, err := routemanager.GetRoutes()
+	if err != nil {
+		return nil, fmt.Errorf("get routes: %w", err)
+	}
+
+	routes := make(map[netip.Prefix]routemanager.Route, len(entries))
+	for _, entry := range entries {
+		routes[entry.Destination] = entry
+	}
+
+	return routes, nil
+}
+
+func stateFromInt(state uint8) string {
+	switch state {
+	case unreachable:
+		return "unreachable"
+	case incomplete:
+		return "incomplete"
+	case probe:
+		return "probe"
+	case delay:
+		return "delay"
+	case stale:
+		return "stale"
+	case reachable:
+		return "reachable"
+	case permanent:
+		return "permanent"
+	case tbd:
+		return "tbd"
+	default:
+		return "unknown"
+	}
+}
+
+func compareIntf(a, b *net.Interface) int {
+	if a == nil && b == nil {
+		return 0
+	}
+	if a == nil {
+		return -1
+	}
+	if b == nil {
+		return 1
+	}
+	return a.Index - b.Index
+}

client/internal/peer/conn.go

@@ -4,6 +4,7 @@ import (
 	"context"
 	"fmt"
 	"net"
+	"net/netip"
 	"runtime"
 	"strings"
 	"sync"
@@ -18,14 +19,17 @@ import (
 	"github.com/netbirdio/netbird/client/internal/wgproxy"
 	"github.com/netbirdio/netbird/iface"
 	"github.com/netbirdio/netbird/iface/bind"
-	signal "github.com/netbirdio/netbird/signal/client"
+	"github.com/netbirdio/netbird/route"
 	sProto "github.com/netbirdio/netbird/signal/proto"
+	nbnet "github.com/netbirdio/netbird/util/net"
 	"github.com/netbirdio/netbird/version"
 )
 
 const (
 	iceKeepAliveDefault           = 4 * time.Second
 	iceDisconnectedTimeoutDefault = 6 * time.Second
+	// iceRelayAcceptanceMinWaitDefault is the same as in the Pion ICE package
+	iceRelayAcceptanceMinWaitDefault = 2 * time.Second
 
 	defaultWgKeepAlive = 25 * time.Second
 )
@@ -65,9 +69,6 @@ type ConnConfig struct {
 
 	NATExternalIPs []string
 
-	// UsesBind indicates whether the WireGuard interface is userspace and uses bind.ICEBind
-	UserspaceBind bool
-
 	// RosenpassPubKey is this peer's Rosenpass public key
 	RosenpassPubKey []byte
 	// RosenpassPubKey is this peer's RosenpassAddr server address (IP:port)
@@ -98,6 +99,9 @@ type IceCredentials struct {
 	Pwd   string
 }
 
+type BeforeAddPeerHookFunc func(connID nbnet.ConnectionID, IP net.IP) error
+type AfterRemovePeerHookFunc func(connID nbnet.ConnectionID) error
+
 type Conn struct {
 	config ConnConfig
 	mu     sync.Mutex
@@ -127,22 +131,13 @@ type Conn struct {
 	wgProxyFactory *wgproxy.Factory
 	wgProxy        wgproxy.Proxy
 
-	remoteModeCh chan ModeMessage
-	meta         meta
+	adapter        iface.TunAdapter
+	iFaceDiscover  stdnet.ExternalIFaceDiscover
+	sentExtraSrflx bool
 
-	adapter       iface.TunAdapter
-	iFaceDiscover stdnet.ExternalIFaceDiscover
-}
-
-// meta holds meta information about a connection
-type meta struct {
-	protoSupport signal.FeaturesSupport
-}
-
-// ModeMessage represents a connection mode chosen by the peer
-type ModeMessage struct {
-	// Direct indicates that it decided to use a direct connection
-	Direct bool
+	connID               nbnet.ConnectionID
+	beforeAddPeerHooks   []BeforeAddPeerHookFunc
+	afterRemovePeerHooks []AfterRemovePeerHookFunc
 }
 
 // GetConf returns the connection config
@@ -171,7 +166,6 @@ func NewConn(config ConnConfig, statusRecorder *Status, wgProxyFactory *wgproxy.
 		remoteOffersCh: make(chan OfferAnswer),
 		remoteAnswerCh: make(chan OfferAnswer),
 		statusRecorder: statusRecorder,
-		remoteModeCh:   make(chan ModeMessage, 1),
 		wgProxyFactory: wgProxyFactory,
 		adapter:        adapter,
 		iFaceDiscover:  iFaceDiscover,
@@ -192,20 +186,22 @@ func (conn *Conn) reCreateAgent() error {
 
 	iceKeepAlive := iceKeepAlive()
 	iceDisconnectedTimeout := iceDisconnectedTimeout()
+	iceRelayAcceptanceMinWait := iceRelayAcceptanceMinWait()
 
 	agentConfig := &ice.AgentConfig{
-		MulticastDNSMode:    ice.MulticastDNSModeDisabled,
-		NetworkTypes:        []ice.NetworkType{ice.NetworkTypeUDP4, ice.NetworkTypeUDP6},
-		Urls:                conn.config.StunTurn,
-		CandidateTypes:      conn.candidateTypes(),
-		FailedTimeout:       &failedTimeout,
-		InterfaceFilter:     stdnet.InterfaceFilter(conn.config.InterfaceBlackList),
-		UDPMux:              conn.config.UDPMux,
-		UDPMuxSrflx:         conn.config.UDPMuxSrflx,
-		NAT1To1IPs:          conn.config.NATExternalIPs,
-		Net:                 transportNet,
-		DisconnectedTimeout: &iceDisconnectedTimeout,
-		KeepaliveInterval:   &iceKeepAlive,
+		MulticastDNSMode:       ice.MulticastDNSModeDisabled,
+		NetworkTypes:           []ice.NetworkType{ice.NetworkTypeUDP4, ice.NetworkTypeUDP6},
+		Urls:                   conn.config.StunTurn,
+		CandidateTypes:         conn.candidateTypes(),
+		FailedTimeout:          &failedTimeout,
+		InterfaceFilter:        stdnet.InterfaceFilter(conn.config.InterfaceBlackList),
+		UDPMux:                 conn.config.UDPMux,
+		UDPMuxSrflx:            conn.config.UDPMuxSrflx,
+		NAT1To1IPs:             conn.config.NATExternalIPs,
+		Net:                    transportNet,
+		DisconnectedTimeout:    &iceDisconnectedTimeout,
+		KeepaliveInterval:      &iceKeepAlive,
+		RelayAcceptanceMinWait: &iceRelayAcceptanceMinWait,
 	}
 
 	if conn.config.DisableIPv6Discovery {
@@ -213,7 +209,6 @@ func (conn *Conn) reCreateAgent() error {
 	}
 
 	conn.agent, err = ice.NewAgent(agentConfig)
-
 	if err != nil {
 		return err
 	}
@@ -233,6 +228,17 @@ func (conn *Conn) reCreateAgent() error {
 		return err
 	}
 
+	err = conn.agent.OnSuccessfulSelectedPairBindingResponse(func(p *ice.CandidatePair) {
+		err := conn.statusRecorder.UpdateLatency(conn.config.Key, p.Latency())
+		if err != nil {
+			log.Debugf("failed to update latency for peer %s: %s", conn.config.Key, err)
+			return
+		}
+	})
+	if err != nil {
+		return fmt.Errorf("failed setting binding response callback: %w", err)
+	}
+
 	return nil
 }
 
@@ -250,7 +256,7 @@ func (conn *Conn) candidateTypes() []ice.CandidateType {
 // Open opens connection to the remote peer starting ICE candidate gathering process.
 // Blocks until connection has been closed or connection timeout.
 // ConnStatus will be set accordingly
-func (conn *Conn) Open() error {
+func (conn *Conn) Open(ctx context.Context) error {
 	log.Debugf("trying to connect to peer %s", conn.config.Key)
 
 	peerState := State{
@@ -258,6 +264,7 @@ func (conn *Conn) Open() error {
 		IP:               strings.Split(conn.config.WgConfig.AllowedIps, "/")[0],
 		ConnStatusUpdate: time.Now(),
 		ConnStatus:       conn.status,
+		Mux:              new(sync.RWMutex),
 	}
 	err := conn.statusRecorder.UpdatePeerState(peerState)
 	if err != nil {
@@ -309,7 +316,7 @@ func (conn *Conn) Open() error {
 	// at this point we received offer/answer and we are ready to gather candidates
 	conn.mu.Lock()
 	conn.status = StatusConnecting
-	conn.ctx, conn.notifyDisconnected = context.WithCancel(context.Background())
+	conn.ctx, conn.notifyDisconnected = context.WithCancel(ctx)
 	defer conn.notifyDisconnected()
 	conn.mu.Unlock()
 
@@ -317,6 +324,7 @@ func (conn *Conn) Open() error {
 		PubKey:           conn.config.Key,
 		ConnStatus:       conn.status,
 		ConnStatusUpdate: time.Now(),
+		Mux:              new(sync.RWMutex),
 	}
 	err = conn.statusRecorder.UpdatePeerState(peerState)
 	if err != nil {
@@ -325,7 +333,7 @@ func (conn *Conn) Open() error {
 
 	err = conn.agent.GatherCandidates()
 	if err != nil {
-		return err
+		return fmt.Errorf("gather candidates: %v", err)
 	}
 
 	// will block until connection succeeded
@@ -342,11 +350,12 @@ func (conn *Conn) Open() error {
 		return err
 	}
 
-	// dynamically set remote WireGuard port is other side specified a different one from the default one
+	// dynamically set remote WireGuard port if other side specified a different one from the default one
 	remoteWgPort := iface.DefaultWgPort
 	if remoteOfferAnswer.WgListenPort != 0 {
 		remoteWgPort = remoteOfferAnswer.WgListenPort
 	}
+
 	// the ice connection has been established successfully so we are ready to start the proxy
 	remoteAddr, err := conn.configureConnection(remoteConn, remoteWgPort, remoteOfferAnswer.RosenpassPubKey,
 		remoteOfferAnswer.RosenpassAddr)
@@ -371,6 +380,14 @@ func isRelayCandidate(candidate ice.Candidate) bool {
 	return candidate.Type() == ice.CandidateTypeRelay
 }
 
+func (conn *Conn) AddBeforeAddPeerHook(hook BeforeAddPeerHookFunc) {
+	conn.beforeAddPeerHooks = append(conn.beforeAddPeerHooks, hook)
+}
+
+func (conn *Conn) AddAfterRemovePeerHook(hook AfterRemovePeerHookFunc) {
+	conn.afterRemovePeerHooks = append(conn.afterRemovePeerHooks, hook)
+}
+
 // configureConnection starts proxying traffic from/to local Wireguard and sets connection status to StatusConnected
 func (conn *Conn) configureConnection(remoteConn net.Conn, remoteWgPort int, remoteRosenpassPubKey []byte, remoteRosenpassAddr string) (net.Addr, error) {
 	conn.mu.Lock()
@@ -384,7 +401,7 @@ func (conn *Conn) configureConnection(remoteConn net.Conn, remoteWgPort int, rem
 	var endpoint net.Addr
 	if isRelayCandidate(pair.Local) {
 		log.Debugf("setup relay connection")
-		conn.wgProxy = conn.wgProxyFactory.GetProxy()
+		conn.wgProxy = conn.wgProxyFactory.GetProxy(conn.ctx)
 		endpoint, err = conn.wgProxy.AddTurnConn(remoteConn)
 		if err != nil {
 			return nil, err
@@ -396,16 +413,30 @@ func (conn *Conn) configureConnection(remoteConn net.Conn, remoteWgPort int, rem
 	}
 
 	endpointUdpAddr, _ := net.ResolveUDPAddr(endpoint.Network(), endpoint.String())
+	log.Debugf("Conn resolved IP for %s: %s", endpoint, endpointUdpAddr.IP)
+
+	conn.connID = nbnet.GenerateConnID()
+	for _, hook := range conn.beforeAddPeerHooks {
+		if err := hook(conn.connID, endpointUdpAddr.IP); err != nil {
+			log.Errorf("Before add peer hook failed: %v", err)
+		}
+	}
 
 	err = conn.config.WgConfig.WgInterface.UpdatePeer(conn.config.WgConfig.RemoteKey, conn.config.WgConfig.AllowedIps, defaultWgKeepAlive, endpointUdpAddr, conn.config.WgConfig.PreSharedKey)
 	if err != nil {
 		if conn.wgProxy != nil {
-			_ = conn.wgProxy.CloseConn()
+			if err := conn.wgProxy.CloseConn(); err != nil {
+				log.Warnf("Failed to close turn connection: %v", err)
+			}
 		}
-		return nil, err
+		return nil, fmt.Errorf("update peer: %w", err)
 	}
 
 	conn.status = StatusConnected
+	rosenpassEnabled := false
+	if remoteRosenpassPubKey != nil {
+		rosenpassEnabled = true
+	}
 
 	peerState := State{
 		PubKey:                     conn.config.Key,
@@ -414,8 +445,10 @@ func (conn *Conn) configureConnection(remoteConn net.Conn, remoteWgPort int, rem
 		LocalIceCandidateType:      pair.Local.Type().String(),
 		RemoteIceCandidateType:     pair.Remote.Type().String(),
 		LocalIceCandidateEndpoint:  fmt.Sprintf("%s:%d", pair.Local.Address(), pair.Local.Port()),
-		RemoteIceCandidateEndpoint: fmt.Sprintf("%s:%d", pair.Remote.Address(), pair.Local.Port()),
+		RemoteIceCandidateEndpoint: fmt.Sprintf("%s:%d", pair.Remote.Address(), pair.Remote.Port()),
 		Direct:                     !isRelayCandidate(pair.Local),
+		RosenpassEnabled:           rosenpassEnabled,
+		Mux:                        new(sync.RWMutex),
 	}
 	if pair.Local.Type() == ice.CandidateTypeRelay || pair.Remote.Type() == ice.CandidateTypeRelay {
 		peerState.Relayed = true
@@ -431,6 +464,10 @@ func (conn *Conn) configureConnection(remoteConn net.Conn, remoteWgPort int, rem
 		return nil, err
 	}
 
+	if runtime.GOOS == "ios" {
+		runtime.GC()
+	}
+
 	if conn.onConnected != nil {
 		conn.onConnected(conn.config.Key, remoteRosenpassPubKey, ipNet.IP.String(), remoteRosenpassAddr)
 	}
@@ -464,6 +501,8 @@ func (conn *Conn) cleanup() error {
 	conn.mu.Lock()
 	defer conn.mu.Unlock()
 
+	conn.sentExtraSrflx = false
+
 	var err1, err2, err3 error
 	if conn.agent != nil {
 		err1 = conn.agent.Close()
@@ -480,6 +519,15 @@ func (conn *Conn) cleanup() error {
 	// todo: is it problem if we try to remove a peer what is never existed?
 	err3 = conn.config.WgConfig.WgInterface.RemovePeer(conn.config.WgConfig.RemoteKey)
 
+	if conn.connID != "" {
+		for _, hook := range conn.afterRemovePeerHooks {
+			if err := hook(conn.connID); err != nil {
+				log.Errorf("After remove peer hook failed: %v", err)
+			}
+		}
+	}
+	conn.connID = ""
+
 	if conn.notifyDisconnected != nil {
 		conn.notifyDisconnected()
 		conn.notifyDisconnected = nil
@@ -495,6 +543,7 @@ func (conn *Conn) cleanup() error {
 		PubKey:           conn.config.Key,
 		ConnStatus:       conn.status,
 		ConnStatusUpdate: time.Now(),
+		Mux:              new(sync.RWMutex),
 	}
 	err := conn.statusRecorder.UpdatePeerState(peerState)
 	if err != nil {
@@ -502,7 +551,7 @@ func (conn *Conn) cleanup() error {
 		// todo rethink status updates
 		log.Debugf("error while updating peer's %s state, err: %v", conn.config.Key, err)
 	}
-	if err := conn.statusRecorder.UpdateWireguardPeerState(conn.config.Key, iface.WGStats{}); err != nil {
+	if err := conn.statusRecorder.UpdateWireGuardPeerState(conn.config.Key, iface.WGStats{}); err != nil {
 		log.Debugf("failed to reset wireguard stats for peer %s: %s", conn.config.Key, err)
 	}
 
@@ -549,16 +598,39 @@ func (conn *Conn) SetSendSignalMessage(handler func(message *sProto.Message) err
 // onICECandidate is a callback attached to an ICE Agent to receive new local connection candidates
 // and then signals them to the remote peer
 func (conn *Conn) onICECandidate(candidate ice.Candidate) {
-	if candidate != nil {
-		// TODO: reported port is incorrect for CandidateTypeHost, makes understanding ICE use via logs confusing as port is ignored
-		log.Debugf("discovered local candidate %s", candidate.String())
-		go func() {
-			err := conn.signalCandidate(candidate)
-			if err != nil {
-				log.Errorf("failed signaling candidate to the remote peer %s %s", conn.config.Key, err)
-			}
-		}()
+	// nil means candidate gathering has been ended
+	if candidate == nil {
+		return
 	}
+
+	// TODO: reported port is incorrect for CandidateTypeHost, makes understanding ICE use via logs confusing as port is ignored
+	log.Debugf("discovered local candidate %s", candidate.String())
+	go func() {
+		err := conn.signalCandidate(candidate)
+		if err != nil {
+			log.Errorf("failed signaling candidate to the remote peer %s %s", conn.config.Key, err)
+		}
+	}()
+
+	if !conn.shouldSendExtraSrflxCandidate(candidate) {
+		return
+	}
+
+	// sends an extra server reflexive candidate to the remote peer with our related port (usually the wireguard port)
+	// this is useful when network has an existing port forwarding rule for the wireguard port and this peer
+	extraSrflx, err := extraSrflxCandidate(candidate)
+	if err != nil {
+		log.Errorf("failed creating extra server reflexive candidate %s", err)
+		return
+	}
+	conn.sentExtraSrflx = true
+
+	go func() {
+		err = conn.signalCandidate(extraSrflx)
+		if err != nil {
+			log.Errorf("failed signaling the extra server reflexive candidate to the remote peer %s: %s", conn.config.Key, err)
+		}
+	}()
 }
 
 func (conn *Conn) onICESelectedCandidatePair(c1 ice.Candidate, c2 ice.Candidate) {
@@ -640,7 +712,7 @@ func (conn *Conn) Close() error {
 		// before conn.Open() another update from management arrives with peers: [1,2,3,4,5]
 		// engine adds a new Conn for 4 and 5
 		// therefore peer 4 has 2 Conn objects
-		log.Warnf("connection has been already closed or attempted closing not started coonection %s", conn.config.Key)
+		log.Warnf("Connection has been already closed or attempted closing not started connection %s", conn.config.Key)
 		return NewConnectionAlreadyClosed(conn.config.Key)
 	}
 }
@@ -683,7 +755,7 @@ func (conn *Conn) OnRemoteAnswer(answer OfferAnswer) bool {
 }
 
 // OnRemoteCandidate Handles ICE connection Candidate provided by the remote peer.
-func (conn *Conn) OnRemoteCandidate(candidate ice.Candidate) {
+func (conn *Conn) OnRemoteCandidate(candidate ice.Candidate, haRoutes route.HAMap) {
 	log.Debugf("OnRemoteCandidate from peer %s -> %s", conn.config.Key, candidate.String())
 	go func() {
 		conn.mu.Lock()
@@ -693,6 +765,10 @@ func (conn *Conn) OnRemoteCandidate(candidate ice.Candidate) {
 			return
 		}
 
+		if candidateViaRoutes(candidate, haRoutes) {
+			return
+		}
+
 		err := conn.agent.AddRemoteCandidate(candidate)
 		if err != nil {
 			log.Errorf("error while handling remote candidate from peer %s", conn.config.Key)
@@ -705,8 +781,49 @@ func (conn *Conn) GetKey() string {
 	return conn.config.Key
 }
 
-// RegisterProtoSupportMeta register supported proto message in the connection metadata
-func (conn *Conn) RegisterProtoSupportMeta(support []uint32) {
-	protoSupport := signal.ParseFeaturesSupported(support)
-	conn.meta.protoSupport = protoSupport
+func (conn *Conn) shouldSendExtraSrflxCandidate(candidate ice.Candidate) bool {
+	if !conn.sentExtraSrflx && candidate.Type() == ice.CandidateTypeServerReflexive && candidate.Port() != candidate.RelatedAddress().Port {
+		return true
+	}
+	return false
+}
+
+func extraSrflxCandidate(candidate ice.Candidate) (*ice.CandidateServerReflexive, error) {
+	relatedAdd := candidate.RelatedAddress()
+	return ice.NewCandidateServerReflexive(&ice.CandidateServerReflexiveConfig{
+		Network:   candidate.NetworkType().String(),
+		Address:   candidate.Address(),
+		Port:      relatedAdd.Port,
+		Component: candidate.Component(),
+		RelAddr:   relatedAdd.Address,
+		RelPort:   relatedAdd.Port,
+	})
+}
+
+func candidateViaRoutes(candidate ice.Candidate, clientRoutes route.HAMap) bool {
+	var routePrefixes []netip.Prefix
+	for _, routes := range clientRoutes {
+		if len(routes) > 0 && routes[0] != nil {
+			routePrefixes = append(routePrefixes, routes[0].Network)
+		}
+	}
+
+	addr, err := netip.ParseAddr(candidate.Address())
+	if err != nil {
+		log.Errorf("Failed to parse IP address %s: %v", candidate.Address(), err)
+		return false
+	}
+
+	for _, prefix := range routePrefixes {
+		// default route is
+		if prefix.Bits() == 0 {
+			continue
+		}
+
+		if prefix.Contains(addr) {
+			log.Debugf("Ignoring candidate [%s], its address is part of routed network %s", candidate.String(), prefix)
+			return true
+		}
+	}
+	return false
 }

client/internal/peer/conn_test.go

@@ -1,6 +1,7 @@
 package peer
 
 import (
+	"context"
 	"sync"
 	"testing"
 	"time"
@@ -35,7 +36,7 @@ func TestNewConn_interfaceFilter(t *testing.T) {
 }
 
 func TestConn_GetKey(t *testing.T) {
-	wgProxyFactory := wgproxy.NewFactory(connConf.LocalWgPort)
+	wgProxyFactory := wgproxy.NewFactory(context.Background(), connConf.LocalWgPort)
 	defer func() {
 		_ = wgProxyFactory.Free()
 	}()
@@ -50,7 +51,7 @@ func TestConn_GetKey(t *testing.T) {
 }
 
 func TestConn_OnRemoteOffer(t *testing.T) {
-	wgProxyFactory := wgproxy.NewFactory(connConf.LocalWgPort)
+	wgProxyFactory := wgproxy.NewFactory(context.Background(), connConf.LocalWgPort)
 	defer func() {
 		_ = wgProxyFactory.Free()
 	}()
@@ -87,7 +88,7 @@ func TestConn_OnRemoteOffer(t *testing.T) {
 }
 
 func TestConn_OnRemoteAnswer(t *testing.T) {
-	wgProxyFactory := wgproxy.NewFactory(connConf.LocalWgPort)
+	wgProxyFactory := wgproxy.NewFactory(context.Background(), connConf.LocalWgPort)
 	defer func() {
 		_ = wgProxyFactory.Free()
 	}()
@@ -123,7 +124,7 @@ func TestConn_OnRemoteAnswer(t *testing.T) {
 	wg.Wait()
 }
 func TestConn_Status(t *testing.T) {
-	wgProxyFactory := wgproxy.NewFactory(connConf.LocalWgPort)
+	wgProxyFactory := wgproxy.NewFactory(context.Background(), connConf.LocalWgPort)
 	defer func() {
 		_ = wgProxyFactory.Free()
 	}()
@@ -153,7 +154,7 @@ func TestConn_Status(t *testing.T) {
 }
 
 func TestConn_Close(t *testing.T) {
-	wgProxyFactory := wgproxy.NewFactory(connConf.LocalWgPort)
+	wgProxyFactory := wgproxy.NewFactory(context.Background(), connConf.LocalWgPort)
 	defer func() {
 		_ = wgProxyFactory.Free()
 	}()

client/internal/peer/env_config.go

@@ -10,9 +10,10 @@ import (
 )
 
 const (
-	envICEKeepAliveIntervalSec   = "NB_ICE_KEEP_ALIVE_INTERVAL_SEC"
-	envICEDisconnectedTimeoutSec = "NB_ICE_DISCONNECTED_TIMEOUT_SEC"
-	envICEForceRelayConn         = "NB_ICE_FORCE_RELAY_CONN"
+	envICEKeepAliveIntervalSec      = "NB_ICE_KEEP_ALIVE_INTERVAL_SEC"
+	envICEDisconnectedTimeoutSec    = "NB_ICE_DISCONNECTED_TIMEOUT_SEC"
+	envICERelayAcceptanceMinWaitSec = "NB_ICE_RELAY_ACCEPTANCE_MIN_WAIT_SEC"
+	envICEForceRelayConn            = "NB_ICE_FORCE_RELAY_CONN"
 )
 
 func iceKeepAlive() time.Duration {
@@ -21,7 +22,7 @@ func iceKeepAlive() time.Duration {
 		return iceKeepAliveDefault
 	}
 
-	log.Debugf("setting ICE keep alive interval to %s seconds", keepAliveEnv)
+	log.Infof("setting ICE keep alive interval to %s seconds", keepAliveEnv)
 	keepAliveEnvSec, err := strconv.Atoi(keepAliveEnv)
 	if err != nil {
 		log.Warnf("invalid value %s set for %s, using default %v", keepAliveEnv, envICEKeepAliveIntervalSec, iceKeepAliveDefault)
@@ -37,7 +38,7 @@ func iceDisconnectedTimeout() time.Duration {
 		return iceDisconnectedTimeoutDefault
 	}
 
-	log.Debugf("setting ICE disconnected timeout to %s seconds", disconnectedTimeoutEnv)
+	log.Infof("setting ICE disconnected timeout to %s seconds", disconnectedTimeoutEnv)
 	disconnectedTimeoutSec, err := strconv.Atoi(disconnectedTimeoutEnv)
 	if err != nil {
 		log.Warnf("invalid value %s set for %s, using default %v", disconnectedTimeoutEnv, envICEDisconnectedTimeoutSec, iceDisconnectedTimeoutDefault)
@@ -47,6 +48,22 @@ func iceDisconnectedTimeout() time.Duration {
 	return time.Duration(disconnectedTimeoutSec) * time.Second
 }
 
+func iceRelayAcceptanceMinWait() time.Duration {
+	iceRelayAcceptanceMinWaitEnv := os.Getenv(envICERelayAcceptanceMinWaitSec)
+	if iceRelayAcceptanceMinWaitEnv == "" {
+		return iceRelayAcceptanceMinWaitDefault
+	}
+
+	log.Infof("setting ICE relay acceptance min wait to %s seconds", iceRelayAcceptanceMinWaitEnv)
+	disconnectedTimeoutSec, err := strconv.Atoi(iceRelayAcceptanceMinWaitEnv)
+	if err != nil {
+		log.Warnf("invalid value %s set for %s, using default %v", iceRelayAcceptanceMinWaitEnv, envICERelayAcceptanceMinWaitSec, iceRelayAcceptanceMinWaitDefault)
+		return iceRelayAcceptanceMinWaitDefault
+	}
+
+	return time.Duration(disconnectedTimeoutSec) * time.Second
+}
+
 func hasICEForceRelayConn() bool {
 	disconnectedTimeoutEnv := os.Getenv(envICEForceRelayConn)
 	return strings.ToLower(disconnectedTimeoutEnv) == "true"

client/internal/peer/status.go

@@ -5,12 +5,16 @@ import (
 	"sync"
 	"time"
 
+	"google.golang.org/grpc/codes"
+	gstatus "google.golang.org/grpc/status"
+
 	"github.com/netbirdio/netbird/client/internal/relay"
 	"github.com/netbirdio/netbird/iface"
 )
 
 // State contains the latest state of a peer
 type State struct {
+	Mux                        *sync.RWMutex
 	IP                         string
 	PubKey                     string
 	FQDN                       string
@@ -25,6 +29,40 @@ type State struct {
 	LastWireguardHandshake     time.Time
 	BytesTx                    int64
 	BytesRx                    int64
+	Latency                    time.Duration
+	RosenpassEnabled           bool
+	routes                     map[string]struct{}
+}
+
+// AddRoute add a single route to routes map
+func (s *State) AddRoute(network string) {
+	s.Mux.Lock()
+	if s.routes == nil {
+		s.routes = make(map[string]struct{})
+	}
+	s.routes[network] = struct{}{}
+	s.Mux.Unlock()
+}
+
+// SetRoutes set state routes
+func (s *State) SetRoutes(routes map[string]struct{}) {
+	s.Mux.Lock()
+	s.routes = routes
+	s.Mux.Unlock()
+}
+
+// DeleteRoute removes a route from the network amp
+func (s *State) DeleteRoute(network string) {
+	s.Mux.Lock()
+	delete(s.routes, network)
+	s.Mux.Unlock()
+}
+
+// GetRoutes return routes map
+func (s *State) GetRoutes() map[string]struct{} {
+	s.Mux.RLock()
+	defer s.Mux.RUnlock()
+	return s.routes
 }
 
 // LocalPeerState contains the latest state of the local peer
@@ -33,6 +71,7 @@ type LocalPeerState struct {
 	PubKey          string
 	KernelInterface bool
 	FQDN            string
+	Routes          map[string]struct{}
 }
 
 // SignalState contains the latest state of a signal connection
@@ -49,30 +88,51 @@ type ManagementState struct {
 	Error     error
 }
 
+// RosenpassState contains the latest state of the Rosenpass configuration
+type RosenpassState struct {
+	Enabled    bool
+	Permissive bool
+}
+
+// NSGroupState represents the status of a DNS server group, including associated domains,
+// whether it's enabled, and the last error message encountered during probing.
+type NSGroupState struct {
+	ID      string
+	Servers []string
+	Domains []string
+	Enabled bool
+	Error   error
+}
+
 // FullStatus contains the full state held by the Status instance
 type FullStatus struct {
 	Peers           []State
 	ManagementState ManagementState
 	SignalState     SignalState
 	LocalPeerState  LocalPeerState
+	RosenpassState  RosenpassState
 	Relays          []relay.ProbeResult
+	NSGroupStates   []NSGroupState
 }
 
 // Status holds a state of peers, signal, management connections and relays
 type Status struct {
-	mux             sync.Mutex
-	peers           map[string]State
-	changeNotify    map[string]chan struct{}
-	signalState     bool
-	signalError     error
-	managementState bool
-	managementError error
-	relayStates     []relay.ProbeResult
-	localPeer       LocalPeerState
-	offlinePeers    []State
-	mgmAddress      string
-	signalAddress   string
-	notifier        *notifier
+	mux                 sync.Mutex
+	peers               map[string]State
+	changeNotify        map[string]chan struct{}
+	signalState         bool
+	signalError         error
+	managementState     bool
+	managementError     error
+	relayStates         []relay.ProbeResult
+	localPeer           LocalPeerState
+	offlinePeers        []State
+	mgmAddress          string
+	signalAddress       string
+	notifier            *notifier
+	rosenpassEnabled    bool
+	rosenpassPermissive bool
+	nsGroupStates       []NSGroupState
 
 	// To reduce the number of notification invocation this bool will be true when need to call the notification
 	// Some Peer actions mostly used by in a batch when the network map has been synchronized. In these type of events
@@ -115,6 +175,7 @@ func (d *Status) AddPeer(peerPubKey string, fqdn string) error {
 		PubKey:     peerPubKey,
 		ConnStatus: StatusDisconnected,
 		FQDN:       fqdn,
+		Mux:        new(sync.RWMutex),
 	}
 	d.peerListChangedForNotification = true
 	return nil
@@ -161,6 +222,10 @@ func (d *Status) UpdatePeerState(receivedState State) error {
 		peerState.IP = receivedState.IP
 	}
 
+	if receivedState.GetRoutes() != nil {
+		peerState.SetRoutes(receivedState.GetRoutes())
+	}
+
 	skipNotification := shouldSkipNotify(receivedState, peerState)
 
 	if receivedState.ConnStatus != peerState.ConnStatus {
@@ -172,6 +237,7 @@ func (d *Status) UpdatePeerState(receivedState State) error {
 		peerState.RemoteIceCandidateType = receivedState.RemoteIceCandidateType
 		peerState.LocalIceCandidateEndpoint = receivedState.LocalIceCandidateEndpoint
 		peerState.RemoteIceCandidateEndpoint = receivedState.RemoteIceCandidateEndpoint
+		peerState.RosenpassEnabled = receivedState.RosenpassEnabled
 	}
 
 	d.peers[receivedState.PubKey] = peerState
@@ -190,8 +256,8 @@ func (d *Status) UpdatePeerState(receivedState State) error {
 	return nil
 }
 
-// UpdateWireguardPeerState updates the wireguard bits of the peer state
-func (d *Status) UpdateWireguardPeerState(pubKey string, wgStats iface.WGStats) error {
+// UpdateWireGuardPeerState updates the WireGuard bits of the peer state
+func (d *Status) UpdateWireGuardPeerState(pubKey string, wgStats iface.WGStats) error {
 	d.mux.Lock()
 	defer d.mux.Unlock()
 
@@ -264,6 +330,13 @@ func (d *Status) GetPeerStateChangeNotifier(peer string) <-chan struct{} {
 	return ch
 }
 
+// GetLocalPeerState returns the local peer state
+func (d *Status) GetLocalPeerState() LocalPeerState {
+	d.mux.Lock()
+	defer d.mux.Unlock()
+	return d.localPeer
+}
+
 // UpdateLocalPeerState updates local peer status
 func (d *Status) UpdateLocalPeerState(localPeerState LocalPeerState) {
 	d.mux.Lock()
@@ -316,6 +389,14 @@ func (d *Status) UpdateManagementAddress(mgmAddress string) {
 	d.mgmAddress = mgmAddress
 }
 
+// UpdateRosenpass update the Rosenpass configuration
+func (d *Status) UpdateRosenpass(rosenpassEnabled, rosenpassPermissive bool) {
+	d.mux.Lock()
+	defer d.mux.Unlock()
+	d.rosenpassPermissive = rosenpassPermissive
+	d.rosenpassEnabled = rosenpassEnabled
+}
+
 // MarkSignalDisconnected sets SignalState to disconnected
 func (d *Status) MarkSignalDisconnected(err error) {
 	d.mux.Lock()
@@ -342,6 +423,19 @@ func (d *Status) UpdateRelayStates(relayResults []relay.ProbeResult) {
 	d.relayStates = relayResults
 }
 
+func (d *Status) UpdateDNSStates(dnsStates []NSGroupState) {
+	d.mux.Lock()
+	defer d.mux.Unlock()
+	d.nsGroupStates = dnsStates
+}
+
+func (d *Status) GetRosenpassState() RosenpassState {
+	return RosenpassState{
+		d.rosenpassEnabled,
+		d.rosenpassPermissive,
+	}
+}
+
 func (d *Status) GetManagementState() ManagementState {
 	return ManagementState{
 		d.mgmAddress,
@@ -350,6 +444,39 @@ func (d *Status) GetManagementState() ManagementState {
 	}
 }
 
+func (d *Status) UpdateLatency(pubKey string, latency time.Duration) error {
+	if latency <= 0 {
+		return nil
+	}
+
+	d.mux.Lock()
+	defer d.mux.Unlock()
+	peerState, ok := d.peers[pubKey]
+	if !ok {
+		return errors.New("peer doesn't exist")
+	}
+	peerState.Latency = latency
+	d.peers[pubKey] = peerState
+	return nil
+}
+
+// IsLoginRequired determines if a peer's login has expired.
+func (d *Status) IsLoginRequired() bool {
+	d.mux.Lock()
+	defer d.mux.Unlock()
+
+	// if peer is connected to the management then login is not expired
+	if d.managementState {
+		return false
+	}
+
+	s, ok := gstatus.FromError(d.managementError)
+	if ok && (s.Code() == codes.InvalidArgument || s.Code() == codes.PermissionDenied) {
+		return true
+	}
+	return false
+}
+
 func (d *Status) GetSignalState() SignalState {
 	return SignalState{
 		d.signalAddress,
@@ -362,6 +489,10 @@ func (d *Status) GetRelayStates() []relay.ProbeResult {
 	return d.relayStates
 }
 
+func (d *Status) GetDNSStates() []NSGroupState {
+	return d.nsGroupStates
+}
+
 // GetFullStatus gets full status
 func (d *Status) GetFullStatus() FullStatus {
 	d.mux.Lock()
@@ -372,6 +503,8 @@ func (d *Status) GetFullStatus() FullStatus {
 		SignalState:     d.GetSignalState(),
 		LocalPeerState:  d.localPeer,
 		Relays:          d.GetRelayStates(),
+		RosenpassState:  d.GetRosenpassState(),
+		NSGroupStates:   d.GetDNSStates(),
 	}
 
 	for _, status := range d.peers {

client/internal/peer/status_test.go

@@ -3,6 +3,7 @@ package peer
 import (
 	"errors"
 	"testing"
+	"sync"
 
 	"github.com/stretchr/testify/assert"
 )
@@ -42,6 +43,7 @@ func TestUpdatePeerState(t *testing.T) {
 	status := NewRecorder("https://mgm")
 	peerState := State{
 		PubKey: key,
+		Mux:	new(sync.RWMutex),
 	}
 
 	status.peers[key] = peerState
@@ -62,6 +64,7 @@ func TestStatus_UpdatePeerFQDN(t *testing.T) {
 	status := NewRecorder("https://mgm")
 	peerState := State{
 		PubKey: key,
+		Mux:	new(sync.RWMutex),
 	}
 
 	status.peers[key] = peerState
@@ -80,6 +83,7 @@ func TestGetPeerStateChangeNotifierLogic(t *testing.T) {
 	status := NewRecorder("https://mgm")
 	peerState := State{
 		PubKey: key,
+		Mux:	new(sync.RWMutex),
 	}
 
 	status.peers[key] = peerState
@@ -104,6 +108,7 @@ func TestRemovePeer(t *testing.T) {
 	status := NewRecorder("https://mgm")
 	peerState := State{
 		PubKey: key,
+		Mux:	new(sync.RWMutex),
 	}
 
 	status.peers[key] = peerState

client/internal/relay/relay.go

@@ -10,6 +10,9 @@ import (
 	"github.com/pion/stun/v2"
 	"github.com/pion/turn/v3"
 	log "github.com/sirupsen/logrus"
+
+	"github.com/netbirdio/netbird/client/internal/stdnet"
+	nbnet "github.com/netbirdio/netbird/util/net"
 )
 
 // ProbeResult holds the info about the result of a relay probe request
@@ -27,7 +30,15 @@ func ProbeSTUN(ctx context.Context, uri *stun.URI) (addr string, probeErr error)
 		}
 	}()
 
-	client, err := stun.DialURI(uri, &stun.DialConfig{})
+	net, err := stdnet.NewNet(nil)
+	if err != nil {
+		probeErr = fmt.Errorf("new net: %w", err)
+		return
+	}
+
+	client, err := stun.DialURI(uri, &stun.DialConfig{
+		Net: net,
+	})
 	if err != nil {
 		probeErr = fmt.Errorf("dial: %w", err)
 		return
@@ -85,14 +96,13 @@ func ProbeTURN(ctx context.Context, uri *stun.URI) (addr string, probeErr error)
 	switch uri.Proto {
 	case stun.ProtoTypeUDP:
 		var err error
-		conn, err = net.ListenPacket("udp", "")
+		conn, err = nbnet.NewListener().ListenPacket(ctx, "udp", "")
 		if err != nil {
 			probeErr = fmt.Errorf("listen: %w", err)
 			return
 		}
 	case stun.ProtoTypeTCP:
-		dialer := net.Dialer{}
-		tcpConn, err := dialer.DialContext(ctx, "tcp", turnServerAddr)
+		tcpConn, err := nbnet.NewDialer().DialContext(ctx, "tcp", turnServerAddr)
 		if err != nil {
 			probeErr = fmt.Errorf("dial: %w", err)
 			return
@@ -109,12 +119,18 @@ func ProbeTURN(ctx context.Context, uri *stun.URI) (addr string, probeErr error)
 		}
 	}()
 
+	net, err := stdnet.NewNet(nil)
+	if err != nil {
+		probeErr = fmt.Errorf("new net: %w", err)
+		return
+	}
 	cfg := &turn.ClientConfig{
 		STUNServerAddr: turnServerAddr,
 		TURNServerAddr: turnServerAddr,
 		Conn:           conn,
 		Username:       uri.Username,
 		Password:       uri.Password,
+		Net:            net,
 	}
 	client, err := turn.NewClient(cfg)
 	if err != nil {
@@ -154,7 +170,7 @@ func ProbeAll(
 
 	var wg sync.WaitGroup
 	for i, uri := range relays {
-		ctx, cancel := context.WithTimeout(ctx, 1*time.Second)
+		ctx, cancel := context.WithTimeout(ctx, 2*time.Second)
 		defer cancel()
 
 		wg.Add(1)

client/internal/routemanager/client.go

@@ -3,7 +3,9 @@ package routemanager
 import (
 	"context"
 	"fmt"
+	"net"
 	"net/netip"
+	"time"
 
 	log "github.com/sirupsen/logrus"
 
@@ -18,6 +20,7 @@ type routerPeerStatus struct {
 	connected bool
 	relayed   bool
 	direct    bool
+	latency   time.Duration
 }
 
 type routesUpdate struct {
@@ -30,7 +33,7 @@ type clientNetwork struct {
 	stop                context.CancelFunc
 	statusRecorder      *peer.Status
 	wgInterface         *iface.WGIface
-	routes              map[string]*route.Route
+	routes              map[route.ID]*route.Route
 	routeUpdate         chan routesUpdate
 	peerStateUpdate     chan struct{}
 	routePeersNotifiers map[string]chan struct{}
@@ -41,12 +44,13 @@ type clientNetwork struct {
 
 func newClientNetworkWatcher(ctx context.Context, wgInterface *iface.WGIface, statusRecorder *peer.Status, network netip.Prefix) *clientNetwork {
 	ctx, cancel := context.WithCancel(ctx)
+
 	client := &clientNetwork{
 		ctx:                 ctx,
 		stop:                cancel,
 		statusRecorder:      statusRecorder,
 		wgInterface:         wgInterface,
-		routes:              make(map[string]*route.Route),
+		routes:              make(map[route.ID]*route.Route),
 		routePeersNotifiers: make(map[string]chan struct{}),
 		routeUpdate:         make(chan routesUpdate),
 		peerStateUpdate:     make(chan struct{}),
@@ -55,8 +59,8 @@ func newClientNetworkWatcher(ctx context.Context, wgInterface *iface.WGIface, st
 	return client
 }
 
-func (c *clientNetwork) getRouterPeerStatuses() map[string]routerPeerStatus {
-	routePeerStatuses := make(map[string]routerPeerStatus)
+func (c *clientNetwork) getRouterPeerStatuses() map[route.ID]routerPeerStatus {
+	routePeerStatuses := make(map[route.ID]routerPeerStatus)
 	for _, r := range c.routes {
 		peerStatus, err := c.statusRecorder.GetPeer(r.Peer)
 		if err != nil {
@@ -67,22 +71,37 @@ func (c *clientNetwork) getRouterPeerStatuses() map[string]routerPeerStatus {
 			connected: peerStatus.ConnStatus == peer.StatusConnected,
 			relayed:   peerStatus.Relayed,
 			direct:    peerStatus.Direct,
+			latency:   peerStatus.Latency,
 		}
 	}
 	return routePeerStatuses
 }
 
-func (c *clientNetwork) getBestRouteFromStatuses(routePeerStatuses map[string]routerPeerStatus) string {
-	chosen := ""
-	chosenScore := 0
+// getBestRouteFromStatuses determines the most optimal route from the available routes
+// within a clientNetwork, taking into account peer connection status, route metrics, and
+// preference for non-relayed and direct connections.
+//
+// It follows these prioritization rules:
+// * Connected peers: Only routes with connected peers are considered.
+// * Metric: Routes with lower metrics (better) are prioritized.
+// * Non-relayed: Routes without relays are preferred.
+// * Direct connections: Routes with direct peer connections are favored.
+// * Stability: In case of equal scores, the currently active route (if any) is maintained.
+// * Latency: Routes with lower latency are prioritized.
+//
+// It returns the ID of the selected optimal route.
+func (c *clientNetwork) getBestRouteFromStatuses(routePeerStatuses map[route.ID]routerPeerStatus) route.ID {
+	chosen := route.ID("")
+	chosenScore := float64(0)
+	currScore := float64(0)
 
-	currID := ""
+	currID := route.ID("")
 	if c.chosenRoute != nil {
 		currID = c.chosenRoute.ID
 	}
 
 	for _, r := range c.routes {
-		tempScore := 0
+		tempScore := float64(0)
 		peerStatus, found := routePeerStatuses[r.ID]
 		if !found || !peerStatus.connected {
 			continue
@@ -90,9 +109,18 @@ func (c *clientNetwork) getBestRouteFromStatuses(routePeerStatuses map[string]ro
 
 		if r.Metric < route.MaxMetric {
 			metricDiff := route.MaxMetric - r.Metric
-			tempScore = metricDiff * 10
+			tempScore = float64(metricDiff) * 10
 		}
 
+		// in some temporal cases, latency can be 0, so we set it to 1s to not block but try to avoid this route
+		latency := time.Second
+		if peerStatus.latency != 0 {
+			latency = peerStatus.latency
+		} else {
+			log.Warnf("peer %s has 0 latency", r.Peer)
+		}
+		tempScore += 1 - latency.Seconds()
+
 		if !peerStatus.relayed {
 			tempScore++
 		}
@@ -101,7 +129,7 @@ func (c *clientNetwork) getBestRouteFromStatuses(routePeerStatuses map[string]ro
 			tempScore++
 		}
 
-		if tempScore > chosenScore || (tempScore == chosenScore && r.ID == currID) {
+		if tempScore > chosenScore || (tempScore == chosenScore && chosen == "") {
 			chosen = r.ID
 			chosenScore = tempScore
 		}
@@ -110,18 +138,31 @@ func (c *clientNetwork) getBestRouteFromStatuses(routePeerStatuses map[string]ro
 			chosen = r.ID
 			chosenScore = tempScore
 		}
+
+		if r.ID == currID {
+			currScore = tempScore
+		}
 	}
 
-	if chosen == "" {
+	switch {
+	case chosen == "":
 		var peers []string
 		for _, r := range c.routes {
 			peers = append(peers, r.Peer)
 		}
 
 		log.Warnf("the network %s has not been assigned a routing peer as no peers from the list %s are currently connected", c.network, peers)
-
-	} else if chosen != currID {
-		log.Infof("new chosen route is %s with peer %s with score %d for network %s", chosen, c.routes[chosen].Peer, chosenScore, c.network)
+	case chosen != currID:
+		// we compare the current score + 10ms to the chosen score to avoid flapping between routes
+		if currScore != 0 && currScore+0.01 > chosenScore {
+			log.Debugf("keeping current routing peer because the score difference with latency is less than 0.01(10ms), current: %f, new: %f", currScore, chosenScore)
+			return currID
+		}
+		var p string
+		if rt := c.routes[chosen]; rt != nil {
+			p = rt.Peer
+		}
+		log.Infof("new chosen route is %s with peer %s with score %f for network %s", chosen, p, chosenScore, c.network)
 	}
 
 	return chosen
@@ -158,15 +199,21 @@ func (c *clientNetwork) startPeersStatusChangeWatcher() {
 func (c *clientNetwork) removeRouteFromWireguardPeer(peerKey string) error {
 	state, err := c.statusRecorder.GetPeer(peerKey)
 	if err != nil {
-		return err
+		return fmt.Errorf("get peer state: %v", err)
 	}
+
+	state.DeleteRoute(c.network.String())
+	if err := c.statusRecorder.UpdatePeerState(state); err != nil {
+		log.Warnf("Failed to update peer state: %v", err)
+	}
+
 	if state.ConnStatus != peer.StatusConnected {
 		return nil
 	}
 
 	err = c.wgInterface.RemoveAllowedIP(peerKey, c.network.String())
 	if err != nil {
-		return fmt.Errorf("couldn't remove allowed IP %s removed for peer %s, err: %v",
+		return fmt.Errorf("remove allowed IP %s removed for peer %s, err: %v",
 			c.network, c.chosenRoute.Peer, err)
 	}
 	return nil
@@ -174,30 +221,26 @@ func (c *clientNetwork) removeRouteFromWireguardPeer(peerKey string) error {
 
 func (c *clientNetwork) removeRouteFromPeerAndSystem() error {
 	if c.chosenRoute != nil {
-		err := c.removeRouteFromWireguardPeer(c.chosenRoute.Peer)
-		if err != nil {
-			return err
+		if err := removeVPNRoute(c.network, c.getAsInterface()); err != nil {
+			return fmt.Errorf("remove route %s from system, err: %v", c.network, err)
 		}
-		err = removeFromRouteTableIfNonSystem(c.network, c.wgInterface.Address().IP.String())
-		if err != nil {
-			return fmt.Errorf("couldn't remove route %s from system, err: %v",
-				c.network, err)
+
+		if err := c.removeRouteFromWireguardPeer(c.chosenRoute.Peer); err != nil {
+			return fmt.Errorf("remove route: %v", err)
 		}
 	}
 	return nil
 }
 
 func (c *clientNetwork) recalculateRouteAndUpdatePeerAndSystem() error {
-
-	var err error
-
 	routerPeerStatuses := c.getRouterPeerStatuses()
 
 	chosen := c.getBestRouteFromStatuses(routerPeerStatuses)
+
+	// If no route is chosen, remove the route from the peer and system
 	if chosen == "" {
-		err = c.removeRouteFromPeerAndSystem()
-		if err != nil {
-			return err
+		if err := c.removeRouteFromPeerAndSystem(); err != nil {
+			return fmt.Errorf("remove route from peer and system: %v", err)
 		}
 
 		c.chosenRoute = nil
@@ -205,6 +248,7 @@ func (c *clientNetwork) recalculateRouteAndUpdatePeerAndSystem() error {
 		return nil
 	}
 
+	// If the chosen route is the same as the current route, do nothing
 	if c.chosenRoute != nil && c.chosenRoute.ID == chosen {
 		if c.chosenRoute.IsEqual(c.routes[chosen]) {
 			return nil
@@ -212,21 +256,31 @@ func (c *clientNetwork) recalculateRouteAndUpdatePeerAndSystem() error {
 	}
 
 	if c.chosenRoute != nil {
-		err = c.removeRouteFromWireguardPeer(c.chosenRoute.Peer)
-		if err != nil {
-			return err
+		// If a previous route exists, remove it from the peer
+		if err := c.removeRouteFromWireguardPeer(c.chosenRoute.Peer); err != nil {
+			return fmt.Errorf("remove route from peer: %v", err)
 		}
 	} else {
-		err = addToRouteTableIfNoExists(c.network, c.wgInterface.Address().IP.String())
-		if err != nil {
+		// otherwise add the route to the system
+		if err := addVPNRoute(c.network, c.getAsInterface()); err != nil {
 			return fmt.Errorf("route %s couldn't be added for peer %s, err: %v",
 				c.network.String(), c.wgInterface.Address().IP.String(), err)
 		}
 	}
 
 	c.chosenRoute = c.routes[chosen]
-	err = c.wgInterface.AddAllowedIP(c.chosenRoute.Peer, c.network.String())
+
+	state, err := c.statusRecorder.GetPeer(c.chosenRoute.Peer)
 	if err != nil {
+		log.Errorf("Failed to get peer state: %v", err)
+	} else {
+		state.AddRoute(c.network.String())
+		if err := c.statusRecorder.UpdatePeerState(state); err != nil {
+			log.Warnf("Failed to update peer state: %v", err)
+		}
+	}
+
+	if err := c.wgInterface.AddAllowedIP(c.chosenRoute.Peer, c.network.String()); err != nil {
 		log.Errorf("couldn't add allowed IP %s added for peer %s, err: %v",
 			c.network, c.chosenRoute.Peer, err)
 	}
@@ -241,7 +295,7 @@ func (c *clientNetwork) sendUpdateToClientNetworkWatcher(update routesUpdate) {
 }
 
 func (c *clientNetwork) handleUpdate(update routesUpdate) {
-	updateMap := make(map[string]*route.Route)
+	updateMap := make(map[route.ID]*route.Route)
 
 	for _, r := range update.routes {
 		updateMap[r.ID] = r
@@ -267,21 +321,21 @@ func (c *clientNetwork) peersStateAndUpdateWatcher() {
 			log.Debugf("stopping watcher for network %s", c.network)
 			err := c.removeRouteFromPeerAndSystem()
 			if err != nil {
-				log.Error(err)
+				log.Errorf("Couldn't remove route from peer and system for network %s: %v", c.network, err)
 			}
 			return
 		case <-c.peerStateUpdate:
 			err := c.recalculateRouteAndUpdatePeerAndSystem()
 			if err != nil {
-				log.Error(err)
+				log.Errorf("Couldn't recalculate route and update peer and system: %v", err)
 			}
 		case update := <-c.routeUpdate:
 			if update.updateSerial < c.updateSerial {
-				log.Warnf("received a routes update with smaller serial number, ignoring it")
+				log.Warnf("Received a routes update with smaller serial number, ignoring it")
 				continue
 			}
 
-			log.Debugf("received a new client network route update for %s", c.network)
+			log.Debugf("Received a new client network route update for %s", c.network)
 
 			c.handleUpdate(update)
 
@@ -289,10 +343,22 @@ func (c *clientNetwork) peersStateAndUpdateWatcher() {
 
 			err := c.recalculateRouteAndUpdatePeerAndSystem()
 			if err != nil {
-				log.Error(err)
+				log.Errorf("Couldn't recalculate route and update peer and system for network %s: %v", c.network, err)
 			}
 
 			c.startPeersStatusChangeWatcher()
 		}
 	}
 }
+
+func (c *clientNetwork) getAsInterface() *net.Interface {
+	intf, err := net.InterfaceByName(c.wgInterface.Name())
+	if err != nil {
+		log.Warnf("Couldn't get interface by name %s: %v", c.wgInterface.Name(), err)
+		intf = &net.Interface{
+			Name: c.wgInterface.Name(),
+		}
+	}
+
+	return intf
+}

client/internal/routemanager/client_test.go

@@ -3,6 +3,7 @@ package routemanager
 import (
 	"net/netip"
 	"testing"
+	"time"
 
 	"github.com/netbirdio/netbird/route"
 )
@@ -11,90 +12,90 @@ func TestGetBestrouteFromStatuses(t *testing.T) {
 
 	testCases := []struct {
 		name            string
-		statuses        map[string]routerPeerStatus
-		expectedRouteID string
-		currentRoute    *route.Route
-		existingRoutes  map[string]*route.Route
+		statuses        map[route.ID]routerPeerStatus
+		expectedRouteID route.ID
+		currentRoute    route.ID
+		existingRoutes  map[route.ID]*route.Route
 	}{
 		{
 			name: "one route",
-			statuses: map[string]routerPeerStatus{
+			statuses: map[route.ID]routerPeerStatus{
 				"route1": {
 					connected: true,
 					relayed:   false,
 					direct:    true,
 				},
 			},
-			existingRoutes: map[string]*route.Route{
+			existingRoutes: map[route.ID]*route.Route{
 				"route1": {
 					ID:     "route1",
 					Metric: route.MaxMetric,
 					Peer:   "peer1",
 				},
 			},
-			currentRoute:    nil,
+			currentRoute:    "",
 			expectedRouteID: "route1",
 		},
 		{
 			name: "one connected routes with relayed and direct",
-			statuses: map[string]routerPeerStatus{
+			statuses: map[route.ID]routerPeerStatus{
 				"route1": {
 					connected: true,
 					relayed:   true,
 					direct:    true,
 				},
 			},
-			existingRoutes: map[string]*route.Route{
+			existingRoutes: map[route.ID]*route.Route{
 				"route1": {
 					ID:     "route1",
 					Metric: route.MaxMetric,
 					Peer:   "peer1",
 				},
 			},
-			currentRoute:    nil,
+			currentRoute:    "",
 			expectedRouteID: "route1",
 		},
 		{
 			name: "one connected routes with relayed and no direct",
-			statuses: map[string]routerPeerStatus{
+			statuses: map[route.ID]routerPeerStatus{
 				"route1": {
 					connected: true,
 					relayed:   true,
 					direct:    false,
 				},
 			},
-			existingRoutes: map[string]*route.Route{
+			existingRoutes: map[route.ID]*route.Route{
 				"route1": {
 					ID:     "route1",
 					Metric: route.MaxMetric,
 					Peer:   "peer1",
 				},
 			},
-			currentRoute:    nil,
+			currentRoute:    "",
 			expectedRouteID: "route1",
 		},
 		{
 			name: "no connected peers",
-			statuses: map[string]routerPeerStatus{
+			statuses: map[route.ID]routerPeerStatus{
 				"route1": {
 					connected: false,
 					relayed:   false,
 					direct:    false,
 				},
 			},
-			existingRoutes: map[string]*route.Route{
+			existingRoutes: map[route.ID]*route.Route{
 				"route1": {
 					ID:     "route1",
 					Metric: route.MaxMetric,
 					Peer:   "peer1",
 				},
 			},
-			currentRoute:    nil,
+			currentRoute:    "",
 			expectedRouteID: "",
 		},
 		{
 			name: "multiple connected peers with different metrics",
-			statuses: map[string]routerPeerStatus{
+			statuses: map[route.ID]routerPeerStatus{
 				"route1": {
 					connected: true,
 					relayed:   false,
@@ -106,7 +107,7 @@ func TestGetBestrouteFromStatuses(t *testing.T) {
 					direct:    true,
 				},
 			},
-			existingRoutes: map[string]*route.Route{
+			existingRoutes: map[route.ID]*route.Route{
 				"route1": {
 					ID:     "route1",
 					Metric: 9000,
@@ -118,12 +119,12 @@ func TestGetBestrouteFromStatuses(t *testing.T) {
 					Peer:   "peer2",
 				},
 			},
-			currentRoute:    nil,
+			currentRoute:    "",
 			expectedRouteID: "route1",
 		},
 		{
 			name: "multiple connected peers with one relayed",
-			statuses: map[string]routerPeerStatus{
+			statuses: map[route.ID]routerPeerStatus{
 				"route1": {
 					connected: true,
 					relayed:   false,
@@ -135,7 +136,7 @@ func TestGetBestrouteFromStatuses(t *testing.T) {
 					direct:    true,
 				},
 			},
-			existingRoutes: map[string]*route.Route{
+			existingRoutes: map[route.ID]*route.Route{
 				"route1": {
 					ID:     "route1",
 					Metric: route.MaxMetric,
@@ -147,12 +148,12 @@ func TestGetBestrouteFromStatuses(t *testing.T) {
 					Peer:   "peer2",
 				},
 			},
-			currentRoute:    nil,
+			currentRoute:    "",
 			expectedRouteID: "route1",
 		},
 		{
 			name: "multiple connected peers with one direct",
-			statuses: map[string]routerPeerStatus{
+			statuses: map[route.ID]routerPeerStatus{
 				"route1": {
 					connected: true,
 					relayed:   false,
@@ -164,7 +165,7 @@ func TestGetBestrouteFromStatuses(t *testing.T) {
 					direct:    false,
 				},
 			},
-			existingRoutes: map[string]*route.Route{
+			existingRoutes: map[route.ID]*route.Route{
 				"route1": {
 					ID:     "route1",
 					Metric: route.MaxMetric,
@@ -176,18 +177,172 @@ func TestGetBestrouteFromStatuses(t *testing.T) {
 					Peer:   "peer2",
 				},
 			},
-			currentRoute:    nil,
+			currentRoute:    "",
 			expectedRouteID: "route1",
 		},
+		{
+			name: "multiple connected peers with different latencies",
+			statuses: map[route.ID]routerPeerStatus{
+				"route1": {
+					connected: true,
+					latency:   300 * time.Millisecond,
+				},
+				"route2": {
+					connected: true,
+					latency:   10 * time.Millisecond,
+				},
+			},
+			existingRoutes: map[route.ID]*route.Route{
+				"route1": {
+					ID:     "route1",
+					Metric: route.MaxMetric,
+					Peer:   "peer1",
+				},
+				"route2": {
+					ID:     "route2",
+					Metric: route.MaxMetric,
+					Peer:   "peer2",
+				},
+			},
+			currentRoute:    "",
+			expectedRouteID: "route2",
+		},
+		{
+			name: "should ignore routes with latency 0",
+			statuses: map[route.ID]routerPeerStatus{
+				"route1": {
+					connected: true,
+					latency:   0 * time.Millisecond,
+				},
+				"route2": {
+					connected: true,
+					latency:   10 * time.Millisecond,
+				},
+			},
+			existingRoutes: map[route.ID]*route.Route{
+				"route1": {
+					ID:     "route1",
+					Metric: route.MaxMetric,
+					Peer:   "peer1",
+				},
+				"route2": {
+					ID:     "route2",
+					Metric: route.MaxMetric,
+					Peer:   "peer2",
+				},
+			},
+			currentRoute:    "",
+			expectedRouteID: "route2",
+		},
+		{
+			name: "current route with similar score and similar but slightly worse latency should not change",
+			statuses: map[route.ID]routerPeerStatus{
+				"route1": {
+					connected: true,
+					relayed:   false,
+					direct:    true,
+					latency:   15 * time.Millisecond,
+				},
+				"route2": {
+					connected: true,
+					relayed:   false,
+					direct:    true,
+					latency:   10 * time.Millisecond,
+				},
+			},
+			existingRoutes: map[route.ID]*route.Route{
+				"route1": {
+					ID:     "route1",
+					Metric: route.MaxMetric,
+					Peer:   "peer1",
+				},
+				"route2": {
+					ID:     "route2",
+					Metric: route.MaxMetric,
+					Peer:   "peer2",
+				},
+			},
+			currentRoute:    "route1",
+			expectedRouteID: "route1",
+		},
+		{
+			name: "current route with bad score should be changed to route with better score",
+			statuses: map[route.ID]routerPeerStatus{
+				"route1": {
+					connected: true,
+					relayed:   false,
+					direct:    true,
+					latency:   200 * time.Millisecond,
+				},
+				"route2": {
+					connected: true,
+					relayed:   false,
+					direct:    true,
+					latency:   10 * time.Millisecond,
+				},
+			},
+			existingRoutes: map[route.ID]*route.Route{
+				"route1": {
+					ID:     "route1",
+					Metric: route.MaxMetric,
+					Peer:   "peer1",
+				},
+				"route2": {
+					ID:     "route2",
+					Metric: route.MaxMetric,
+					Peer:   "peer2",
+				},
+			},
+			currentRoute:    "route1",
+			expectedRouteID: "route2",
+		},
+		{
+			name: "current chosen route doesn't exist anymore",
+			statuses: map[route.ID]routerPeerStatus{
+				"route1": {
+					connected: true,
+					relayed:   false,
+					direct:    true,
+					latency:   20 * time.Millisecond,
+				},
+				"route2": {
+					connected: true,
+					relayed:   false,
+					direct:    true,
+					latency:   10 * time.Millisecond,
+				},
+			},
+			existingRoutes: map[route.ID]*route.Route{
+				"route1": {
+					ID:     "route1",
+					Metric: route.MaxMetric,
+					Peer:   "peer1",
+				},
+				"route2": {
+					ID:     "route2",
+					Metric: route.MaxMetric,
+					Peer:   "peer2",
+				},
+			},
+			currentRoute:    "routeDoesntExistAnymore",
+			expectedRouteID: "route2",
+		},
 	}
 
 	for _, tc := range testCases {
 		t.Run(tc.name, func(t *testing.T) {
+			currentRoute := &route.Route{
+				ID: "routeDoesntExistAnymore",
+			}
+			if tc.currentRoute != "" {
+				currentRoute = tc.existingRoutes[tc.currentRoute]
+			}
+
 			// create new clientNetwork
 			client := &clientNetwork{
 				network:     netip.MustParsePrefix("192.168.0.0/24"),
 				routes:      tc.existingRoutes,
-				chosenRoute: tc.currentRoute,
+				chosenRoute: currentRoute,
 			}
 
 			chosenRoute := client.getBestRouteFromStatuses(tc.statuses)

client/internal/routemanager/manager.go

@@ -2,6 +2,10 @@ package routemanager
 
 import (
 	"context"
+	"fmt"
+	"net"
+	"net/netip"
+	"net/url"
 	"runtime"
 	"sync"
 
@@ -10,14 +14,24 @@ import (
 	firewall "github.com/netbirdio/netbird/client/firewall/manager"
 	"github.com/netbirdio/netbird/client/internal/listener"
 	"github.com/netbirdio/netbird/client/internal/peer"
+	"github.com/netbirdio/netbird/client/internal/routeselector"
 	"github.com/netbirdio/netbird/iface"
 	"github.com/netbirdio/netbird/route"
+	nbnet "github.com/netbirdio/netbird/util/net"
 	"github.com/netbirdio/netbird/version"
 )
 
+var defaultv4 = netip.PrefixFrom(netip.IPv4Unspecified(), 0)
+
+// nolint:unused
+var defaultv6 = netip.PrefixFrom(netip.IPv6Unspecified(), 0)
+
 // Manager is a route manager interface
 type Manager interface {
-	UpdateRoutes(updateSerial uint64, newRoutes []*route.Route) error
+	Init() (peer.BeforeAddPeerHookFunc, peer.AfterRemovePeerHookFunc, error)
+	UpdateRoutes(updateSerial uint64, newRoutes []*route.Route) (map[route.ID]*route.Route, route.HAMap, error)
+	TriggerSelection(route.HAMap)
+	GetRouteSelector() *routeselector.RouteSelector
 	SetRouteChangeListener(listener listener.NetworkChangeListener)
 	InitialRouteRange() []string
 	EnableServerRouter(firewall firewall.Manager) error
@@ -29,7 +43,8 @@ type DefaultManager struct {
 	ctx            context.Context
 	stop           context.CancelFunc
 	mux            sync.Mutex
-	clientNetworks map[string]*clientNetwork
+	clientNetworks map[route.HAUniqueID]*clientNetwork
+	routeSelector  *routeselector.RouteSelector
 	serverRouter   serverRouter
 	statusRecorder *peer.Status
 	wgInterface    *iface.WGIface
@@ -42,7 +57,8 @@ func NewManager(ctx context.Context, pubKey string, wgInterface *iface.WGIface,
 	dm := &DefaultManager{
 		ctx:            mCTX,
 		stop:           cancel,
-		clientNetworks: make(map[string]*clientNetwork),
+		clientNetworks: make(map[route.HAUniqueID]*clientNetwork),
+		routeSelector:  routeselector.NewRouteSelector(),
 		statusRecorder: statusRecorder,
 		wgInterface:    wgInterface,
 		pubKey:         pubKey,
@@ -56,9 +72,31 @@ func NewManager(ctx context.Context, pubKey string, wgInterface *iface.WGIface,
 	return dm
 }
 
+// Init sets up the routing
+func (m *DefaultManager) Init() (peer.BeforeAddPeerHookFunc, peer.AfterRemovePeerHookFunc, error) {
+	if nbnet.CustomRoutingDisabled() {
+		return nil, nil, nil
+	}
+
+	if err := cleanupRouting(); err != nil {
+		log.Warnf("Failed cleaning up routing: %v", err)
+	}
+
+	mgmtAddress := m.statusRecorder.GetManagementState().URL
+	signalAddress := m.statusRecorder.GetSignalState().URL
+	ips := resolveURLsToIPs([]string{mgmtAddress, signalAddress})
+
+	beforePeerHook, afterPeerHook, err := setupRouting(ips, m.wgInterface)
+	if err != nil {
+		return nil, nil, fmt.Errorf("setup routing: %w", err)
+	}
+	log.Info("Routing setup complete")
+	return beforePeerHook, afterPeerHook, nil
+}
+
 func (m *DefaultManager) EnableServerRouter(firewall firewall.Manager) error {
 	var err error
-	m.serverRouter, err = newServerRouter(m.ctx, m.wgInterface, firewall)
+	m.serverRouter, err = newServerRouter(m.ctx, m.wgInterface, firewall, m.statusRecorder)
 	if err != nil {
 		return err
 	}
@@ -71,32 +109,42 @@ func (m *DefaultManager) Stop() {
 	if m.serverRouter != nil {
 		m.serverRouter.cleanUp()
 	}
+
+	if !nbnet.CustomRoutingDisabled() {
+		if err := cleanupRouting(); err != nil {
+			log.Errorf("Error cleaning up routing: %v", err)
+		} else {
+			log.Info("Routing cleanup complete")
+		}
+	}
+
 	m.ctx = nil
 }
 
-// UpdateRoutes compares received routes with existing routes and remove, update or add them to the client and server maps
-func (m *DefaultManager) UpdateRoutes(updateSerial uint64, newRoutes []*route.Route) error {
+// UpdateRoutes compares received routes with existing routes and removes, updates or adds them to the client and server maps
+func (m *DefaultManager) UpdateRoutes(updateSerial uint64, newRoutes []*route.Route) (map[route.ID]*route.Route, route.HAMap, error) {
 	select {
 	case <-m.ctx.Done():
 		log.Infof("not updating routes as context is closed")
-		return m.ctx.Err()
+		return nil, nil, m.ctx.Err()
 	default:
 		m.mux.Lock()
 		defer m.mux.Unlock()
 
-		newServerRoutesMap, newClientRoutesIDMap := m.classifiesRoutes(newRoutes)
+		newServerRoutesMap, newClientRoutesIDMap := m.classifyRoutes(newRoutes)
 
-		m.updateClientNetworks(updateSerial, newClientRoutesIDMap)
-		m.notifier.onNewRoutes(newClientRoutesIDMap)
+		filteredClientRoutes := m.routeSelector.FilterSelected(newClientRoutesIDMap)
+		m.updateClientNetworks(updateSerial, filteredClientRoutes)
+		m.notifier.onNewRoutes(filteredClientRoutes)
 
 		if m.serverRouter != nil {
 			err := m.serverRouter.updateRoutes(newServerRoutesMap)
 			if err != nil {
-				return err
+				return nil, nil, fmt.Errorf("update routes: %w", err)
 			}
 		}
 
-		return nil
+		return newServerRoutesMap, newClientRoutesIDMap, nil
 	}
 }
 
@@ -107,19 +155,57 @@ func (m *DefaultManager) SetRouteChangeListener(listener listener.NetworkChangeL
 
 // InitialRouteRange return the list of initial routes. It used by mobile systems
 func (m *DefaultManager) InitialRouteRange() []string {
-	return m.notifier.initialRouteRanges()
+	return m.notifier.getInitialRouteRanges()
 }
 
-func (m *DefaultManager) updateClientNetworks(updateSerial uint64, networks map[string][]*route.Route) {
-	// removing routes that do not exist as per the update from the Management service.
+// GetRouteSelector returns the route selector
+func (m *DefaultManager) GetRouteSelector() *routeselector.RouteSelector {
+	return m.routeSelector
+}
+
+// GetClientRoutes returns the client routes
+func (m *DefaultManager) GetClientRoutes() map[route.HAUniqueID]*clientNetwork {
+	return m.clientNetworks
+}
+
+// TriggerSelection triggers the selection of routes, stopping deselected watchers and starting newly selected ones
+func (m *DefaultManager) TriggerSelection(networks route.HAMap) {
+	m.mux.Lock()
+	defer m.mux.Unlock()
+
+	networks = m.routeSelector.FilterSelected(networks)
+
+	m.notifier.onNewRoutes(networks)
+
+	m.stopObsoleteClients(networks)
+
+	for id, routes := range networks {
+		if _, found := m.clientNetworks[id]; found {
+			// don't touch existing client network watchers
+			continue
+		}
+
+		clientNetworkWatcher := newClientNetworkWatcher(m.ctx, m.wgInterface, m.statusRecorder, routes[0].Network)
+		m.clientNetworks[id] = clientNetworkWatcher
+		go clientNetworkWatcher.peersStateAndUpdateWatcher()
+		clientNetworkWatcher.sendUpdateToClientNetworkWatcher(routesUpdate{routes: routes})
+	}
+}
+
+// stopObsoleteClients stops the client network watcher for the networks that are not in the new list
+func (m *DefaultManager) stopObsoleteClients(networks route.HAMap) {
 	for id, client := range m.clientNetworks {
-		_, found := networks[id]
-		if !found {
-			log.Debugf("stopping client network watcher, %s", id)
+		if _, ok := networks[id]; !ok {
+			log.Debugf("Stopping client network watcher, %s", id)
 			client.stop()
 			delete(m.clientNetworks, id)
 		}
 	}
+}
+
+func (m *DefaultManager) updateClientNetworks(updateSerial uint64, networks route.HAMap) {
+	// removing routes that do not exist as per the update from the Management service.
+	m.stopObsoleteClients(networks)
 
 	for id, routes := range networks {
 		clientNetworkWatcher, found := m.clientNetworks[id]
@@ -136,15 +222,15 @@ func (m *DefaultManager) updateClientNetworks(updateSerial uint64, networks map[
 	}
 }
 
-func (m *DefaultManager) classifiesRoutes(newRoutes []*route.Route) (map[string]*route.Route, map[string][]*route.Route) {
-	newClientRoutesIDMap := make(map[string][]*route.Route)
-	newServerRoutesMap := make(map[string]*route.Route)
-	ownNetworkIDs := make(map[string]bool)
+func (m *DefaultManager) classifyRoutes(newRoutes []*route.Route) (map[route.ID]*route.Route, route.HAMap) {
+	newClientRoutesIDMap := make(route.HAMap)
+	newServerRoutesMap := make(map[route.ID]*route.Route)
+	ownNetworkIDs := make(map[route.HAUniqueID]bool)
 
 	for _, newRoute := range newRoutes {
-		networkID := route.GetHAUniqueID(newRoute)
+		haID := route.GetHAUniqueID(newRoute)
 		if newRoute.Peer == m.pubKey {
-			ownNetworkIDs[networkID] = true
+			ownNetworkIDs[haID] = true
 			// only linux is supported for now
 			if runtime.GOOS != "linux" {
 				log.Warnf("received a route to manage, but agent doesn't support router mode on %s OS", runtime.GOOS)
@@ -155,16 +241,12 @@ func (m *DefaultManager) classifiesRoutes(newRoutes []*route.Route) (map[string]
 	}
 
 	for _, newRoute := range newRoutes {
-		networkID := route.GetHAUniqueID(newRoute)
-		if !ownNetworkIDs[networkID] {
-			// if prefix is too small, lets assume is a possible default route which is not yet supported
-			// we skip this route management
-			if newRoute.Network.Bits() < minRangeBits {
-				log.Errorf("this agent version: %s, doesn't support default routes, received %s, skipping this route",
-					version.NetbirdVersion(), newRoute.Network)
+		haID := route.GetHAUniqueID(newRoute)
+		if !ownNetworkIDs[haID] {
+			if !isPrefixSupported(newRoute.Network) {
 				continue
 			}
-			newClientRoutesIDMap[networkID] = append(newClientRoutesIDMap[networkID], newRoute)
+			newClientRoutesIDMap[haID] = append(newClientRoutesIDMap[haID], newRoute)
 		}
 	}
 
@@ -172,10 +254,44 @@ func (m *DefaultManager) classifiesRoutes(newRoutes []*route.Route) (map[string]
 }
 
 func (m *DefaultManager) clientRoutes(initialRoutes []*route.Route) []*route.Route {
-	_, crMap := m.classifiesRoutes(initialRoutes)
+	_, crMap := m.classifyRoutes(initialRoutes)
 	rs := make([]*route.Route, 0)
 	for _, routes := range crMap {
 		rs = append(rs, routes...)
 	}
 	return rs
 }
+
+func isPrefixSupported(prefix netip.Prefix) bool {
+	if !nbnet.CustomRoutingDisabled() {
+		return true
+	}
+
+	// If prefix is too small, lets assume it is a possible default prefix which is not yet supported
+	// we skip this prefix management
+	if prefix.Bits() <= minRangeBits {
+		log.Warnf("This agent version: %s, doesn't support default routes, received %s, skipping this prefix",
+			version.NetbirdVersion(), prefix)
+		return false
+	}
+	return true
+}
+
+// resolveURLsToIPs takes a slice of URLs, resolves them to IP addresses and returns a slice of IPs.
+func resolveURLsToIPs(urls []string) []net.IP {
+	var ips []net.IP
+	for _, rawurl := range urls {
+		u, err := url.Parse(rawurl)
+		if err != nil {
+			log.Errorf("Failed to parse url %s: %v", rawurl, err)
+			continue
+		}
+		ipAddrs, err := net.LookupIP(u.Hostname())
+		if err != nil {
+			log.Errorf("Failed to resolve host %s: %v", u.Hostname(), err)
+			continue
+		}
+		ips = append(ips, ipAddrs...)
+	}
+	return ips
+}

client/internal/routemanager/manager_test.go

@@ -28,13 +28,14 @@ const remotePeerKey2 = "remote1"
 
 func TestManagerUpdateRoutes(t *testing.T) {
 	testCases := []struct {
-		name                          string
-		inputInitRoutes               []*route.Route
-		inputRoutes                   []*route.Route
-		inputSerial                   uint64
-		removeSrvRouter               bool
-		serverRoutesExpected          int
-		clientNetworkWatchersExpected int
+		name                                 string
+		inputInitRoutes                      []*route.Route
+		inputRoutes                          []*route.Route
+		inputSerial                          uint64
+		removeSrvRouter                      bool
+		serverRoutesExpected                 int
+		clientNetworkWatchersExpected        int
+		clientNetworkWatchersExpectedAllowed int
 	}{
 		{
 			name:            "Should create 2 client networks",
@@ -200,8 +201,9 @@ func TestManagerUpdateRoutes(t *testing.T) {
 					Enabled:     true,
 				},
 			},
-			inputSerial:                   1,
-			clientNetworkWatchersExpected: 0,
+			inputSerial:                          1,
+			clientNetworkWatchersExpected:        0,
+			clientNetworkWatchersExpectedAllowed: 1,
 		},
 		{
 			name: "Remove 1 Client Route",
@@ -415,6 +417,10 @@ func TestManagerUpdateRoutes(t *testing.T) {
 			statusRecorder := peer.NewRecorder("https://mgm")
 			ctx := context.TODO()
 			routeManager := NewManager(ctx, localPeerKey, wgInterface, statusRecorder, nil)
+
+			_, _, err = routeManager.Init()
+
+			require.NoError(t, err, "should init route manager")
 			defer routeManager.Stop()
 
 			if testCase.removeSrvRouter {
@@ -422,14 +428,18 @@ func TestManagerUpdateRoutes(t *testing.T) {
 			}
 
 			if len(testCase.inputInitRoutes) > 0 {
-				err = routeManager.UpdateRoutes(testCase.inputSerial, testCase.inputRoutes)
+				_, _, err = routeManager.UpdateRoutes(testCase.inputSerial, testCase.inputRoutes)
 				require.NoError(t, err, "should update routes with init routes")
 			}
 
-			err = routeManager.UpdateRoutes(testCase.inputSerial+uint64(len(testCase.inputInitRoutes)), testCase.inputRoutes)
+			_, _, err = routeManager.UpdateRoutes(testCase.inputSerial+uint64(len(testCase.inputInitRoutes)), testCase.inputRoutes)
 			require.NoError(t, err, "should update routes")
 
-			require.Len(t, routeManager.clientNetworks, testCase.clientNetworkWatchersExpected, "client networks size should match")
+			expectedWatchers := testCase.clientNetworkWatchersExpected
+			if (runtime.GOOS == "linux" || runtime.GOOS == "windows" || runtime.GOOS == "darwin") && testCase.clientNetworkWatchersExpectedAllowed != 0 {
+				expectedWatchers = testCase.clientNetworkWatchersExpectedAllowed
+			}
+			require.Len(t, routeManager.clientNetworks, expectedWatchers, "client networks size should match")
 
 			if runtime.GOOS == "linux" && routeManager.serverRouter != nil {
 				sr := routeManager.serverRouter.(*defaultServerRouter)

client/internal/routemanager/mock.go

@@ -6,14 +6,22 @@ import (
 
 	firewall "github.com/netbirdio/netbird/client/firewall/manager"
 	"github.com/netbirdio/netbird/client/internal/listener"
+	"github.com/netbirdio/netbird/client/internal/peer"
+	"github.com/netbirdio/netbird/client/internal/routeselector"
 	"github.com/netbirdio/netbird/iface"
 	"github.com/netbirdio/netbird/route"
 )
 
 // MockManager is the mock instance of a route manager
 type MockManager struct {
-	UpdateRoutesFunc func(updateSerial uint64, newRoutes []*route.Route) error
-	StopFunc         func()
+	UpdateRoutesFunc     func(updateSerial uint64, newRoutes []*route.Route) (map[route.ID]*route.Route, route.HAMap, error)
+	TriggerSelectionFunc func(haMap route.HAMap)
+	GetRouteSelectorFunc func() *routeselector.RouteSelector
+	StopFunc             func()
+}
+
+func (m *MockManager) Init() (peer.BeforeAddPeerHookFunc, peer.AfterRemovePeerHookFunc, error) {
+	return nil, nil, nil
 }
 
 // InitialRouteRange mock implementation of InitialRouteRange from Manager interface
@@ -22,11 +30,25 @@ func (m *MockManager) InitialRouteRange() []string {
 }
 
 // UpdateRoutes mock implementation of UpdateRoutes from Manager interface
-func (m *MockManager) UpdateRoutes(updateSerial uint64, newRoutes []*route.Route) error {
+func (m *MockManager) UpdateRoutes(updateSerial uint64, newRoutes []*route.Route) (map[route.ID]*route.Route, route.HAMap, error) {
 	if m.UpdateRoutesFunc != nil {
 		return m.UpdateRoutesFunc(updateSerial, newRoutes)
 	}
-	return fmt.Errorf("method UpdateRoutes is not implemented")
+	return nil, nil, fmt.Errorf("method UpdateRoutes is not implemented")
+}
+
+func (m *MockManager) TriggerSelection(networks route.HAMap) {
+	if m.TriggerSelectionFunc != nil {
+		m.TriggerSelectionFunc(networks)
+	}
+}
+
+// GetRouteSelector mock implementation of GetRouteSelector from Manager interface
+func (m *MockManager) GetRouteSelector() *routeselector.RouteSelector {
+	if m.GetRouteSelectorFunc != nil {
+		return m.GetRouteSelectorFunc()
+	}
+	return nil
 }
 
 // Start mock implementation of Start from Manager interface

client/internal/routemanager/notifier.go

@@ -1,6 +1,7 @@
 package routemanager
 
 import (
+	"runtime"
 	"sort"
 	"strings"
 	"sync"
@@ -10,8 +11,8 @@ import (
 )
 
 type notifier struct {
-	initialRouteRangers []string
-	routeRangers        []string
+	initialRouteRanges []string
+	routeRanges        []string
 
 	listener    listener.NetworkChangeListener
 	listenerMux sync.Mutex
@@ -33,10 +34,10 @@ func (n *notifier) setInitialClientRoutes(clientRoutes []*route.Route) {
 		nets = append(nets, r.Network.String())
 	}
 	sort.Strings(nets)
-	n.initialRouteRangers = nets
+	n.initialRouteRanges = nets
 }
 
-func (n *notifier) onNewRoutes(idMap map[string][]*route.Route) {
+func (n *notifier) onNewRoutes(idMap route.HAMap) {
 	newNets := make([]string, 0)
 	for _, routes := range idMap {
 		for _, r := range routes {
@@ -45,11 +46,18 @@ func (n *notifier) onNewRoutes(idMap map[string][]*route.Route) {
 	}
 
 	sort.Strings(newNets)
-	if !n.hasDiff(n.initialRouteRangers, newNets) {
-		return
+	switch runtime.GOOS {
+	case "android":
+		if !n.hasDiff(n.initialRouteRanges, newNets) {
+			return
+		}
+	default:
+		if !n.hasDiff(n.routeRanges, newNets) {
+			return
+		}
 	}
 
-	n.routeRangers = newNets
+	n.routeRanges = newNets
 
 	n.notify()
 }
@@ -62,7 +70,7 @@ func (n *notifier) notify() {
 	}
 
 	go func(l listener.NetworkChangeListener) {
-		l.OnNetworkChanged(strings.Join(n.routeRangers, ","))
+		l.OnNetworkChanged(strings.Join(addIPv6RangeIfNeeded(n.routeRanges), ","))
 	}(n.listener)
 }
 
@@ -78,6 +86,20 @@ func (n *notifier) hasDiff(a []string, b []string) bool {
 	return false
 }
 
-func (n *notifier) initialRouteRanges() []string {
-	return n.initialRouteRangers
+func (n *notifier) getInitialRouteRanges() []string {
+	return addIPv6RangeIfNeeded(n.initialRouteRanges)
+}
+
+// addIPv6RangeIfNeeded returns the input ranges with the default IPv6 range when there is an IPv4 default route.
+func addIPv6RangeIfNeeded(inputRanges []string) []string {
+	ranges := inputRanges
+	for _, r := range inputRanges {
+		// we are intentionally adding the ipv6 default range in case of ipv4 default range
+		// to ensure that all traffic is managed by the tunnel interface on android
+		if r == "0.0.0.0/0" {
+			ranges = append(ranges, "::/0")
+			break
+		}
+	}
+	return ranges
 }

client/internal/routemanager/routemanager.go

@@ -0,0 +1,127 @@
+//go:build !android && !ios
+
+package routemanager
+
+import (
+	"errors"
+	"fmt"
+	"net"
+	"net/netip"
+	"sync"
+
+	"github.com/hashicorp/go-multierror"
+	log "github.com/sirupsen/logrus"
+
+	nbnet "github.com/netbirdio/netbird/util/net"
+)
+
+type ref struct {
+	count   int
+	nexthop netip.Addr
+	intf    *net.Interface
+}
+
+type RouteManager struct {
+	// refCountMap keeps track of the reference ref for prefixes
+	refCountMap map[netip.Prefix]ref
+	// prefixMap keeps track of the prefixes associated with a connection ID for removal
+	prefixMap   map[nbnet.ConnectionID][]netip.Prefix
+	addRoute    AddRouteFunc
+	removeRoute RemoveRouteFunc
+	mutex       sync.Mutex
+}
+
+type AddRouteFunc func(prefix netip.Prefix) (nexthop netip.Addr, intf *net.Interface, err error)
+type RemoveRouteFunc func(prefix netip.Prefix, nexthop netip.Addr, intf *net.Interface) error
+
+func NewRouteManager(addRoute AddRouteFunc, removeRoute RemoveRouteFunc) *RouteManager {
+	// TODO: read initial routing table into refCountMap
+	return &RouteManager{
+		refCountMap: map[netip.Prefix]ref{},
+		prefixMap:   map[nbnet.ConnectionID][]netip.Prefix{},
+		addRoute:    addRoute,
+		removeRoute: removeRoute,
+	}
+}
+
+func (rm *RouteManager) AddRouteRef(connID nbnet.ConnectionID, prefix netip.Prefix) error {
+	rm.mutex.Lock()
+	defer rm.mutex.Unlock()
+
+	ref := rm.refCountMap[prefix]
+	log.Debugf("Increasing route ref count %d for prefix %s", ref.count, prefix)
+
+	// Add route to the system, only if it's a new prefix
+	if ref.count == 0 {
+		log.Debugf("Adding route for prefix %s", prefix)
+		nexthop, intf, err := rm.addRoute(prefix)
+		if errors.Is(err, ErrRouteNotFound) {
+			return nil
+		}
+		if errors.Is(err, ErrRouteNotAllowed) {
+			log.Debugf("Adding route for prefix %s: %s", prefix, err)
+		}
+		if err != nil {
+			return fmt.Errorf("failed to add route for prefix %s: %w", prefix, err)
+		}
+		ref.nexthop = nexthop
+		ref.intf = intf
+	}
+
+	ref.count++
+	rm.refCountMap[prefix] = ref
+	rm.prefixMap[connID] = append(rm.prefixMap[connID], prefix)
+
+	return nil
+}
+
+func (rm *RouteManager) RemoveRouteRef(connID nbnet.ConnectionID) error {
+	rm.mutex.Lock()
+	defer rm.mutex.Unlock()
+
+	prefixes, ok := rm.prefixMap[connID]
+	if !ok {
+		log.Debugf("No prefixes found for connection ID %s", connID)
+		return nil
+	}
+
+	var result *multierror.Error
+	for _, prefix := range prefixes {
+		ref := rm.refCountMap[prefix]
+		log.Debugf("Decreasing route ref count %d for prefix %s", ref.count, prefix)
+		if ref.count == 1 {
+			log.Debugf("Removing route for prefix %s", prefix)
+			// TODO: don't fail if the route is not found
+			if err := rm.removeRoute(prefix, ref.nexthop, ref.intf); err != nil {
+				result = multierror.Append(result, fmt.Errorf("remove route for prefix %s: %w", prefix, err))
+				continue
+			}
+			delete(rm.refCountMap, prefix)
+		} else {
+			ref.count--
+			rm.refCountMap[prefix] = ref
+		}
+	}
+	delete(rm.prefixMap, connID)
+
+	return result.ErrorOrNil()
+}
+
+// Flush removes all references and routes from the system
+func (rm *RouteManager) Flush() error {
+	rm.mutex.Lock()
+	defer rm.mutex.Unlock()
+
+	var result *multierror.Error
+	for prefix := range rm.refCountMap {
+		log.Debugf("Removing route for prefix %s", prefix)
+		ref := rm.refCountMap[prefix]
+		if err := rm.removeRoute(prefix, ref.nexthop, ref.intf); err != nil {
+			result = multierror.Append(result, fmt.Errorf("remove route for prefix %s: %w", prefix, err))
+		}
+	}
+	rm.refCountMap = map[netip.Prefix]ref{}
+	rm.prefixMap = map[nbnet.ConnectionID][]netip.Prefix{}
+
+	return result.ErrorOrNil()
+}

client/internal/routemanager/server.go

@@ -3,7 +3,7 @@ package routemanager
 import "github.com/netbirdio/netbird/route"
 
 type serverRouter interface {
-	updateRoutes(map[string]*route.Route) error
+	updateRoutes(map[route.ID]*route.Route) error
 	removeFromServerNetwork(*route.Route) error
 	cleanUp()
 }