Aligns new tests to signature

by moving the persistence from applySync to the map state manager

.github/workflows/check-license-dependencies.yml

@@ -20,7 +20,7 @@ jobs:
 
     steps:
       - name: Checkout code
-        uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2
+        uses: actions/checkout@9c091bb21b7c1c1d1991bb908d89e4e9dddfe3e0 # v7.0.0
         with:
           persist-credentials: false
 
@@ -59,12 +59,12 @@ jobs:
     runs-on: ubuntu-latest
 
     steps:
-      - uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2
+      - uses: actions/checkout@9c091bb21b7c1c1d1991bb908d89e4e9dddfe3e0 # v7.0.0
         with:
           persist-credentials: false
 
       - name: Set up Go
-        uses: actions/setup-go@4b73464bb391d4059bd26b0524d20df3927bd417 # v6.3.0
+        uses: actions/setup-go@924ae3a1cded613372ab5595356fb5720e22ba16 # v6.5.0
         with:
           go-version-file: "go.mod"
           cache: true

.github/workflows/git-town.yml

@@ -15,7 +15,7 @@ jobs:
       pull-requests: write
 
     steps:
-      - uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2
+      - uses: actions/checkout@9c091bb21b7c1c1d1991bb908d89e4e9dddfe3e0 # v7.0.0
         with:
           persist-credentials: false
       - uses: git-town/action@3d8b878379abb1ee393fb49865a28b4a6c2cd3b0 # v1.2.1

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

@@ -16,18 +16,18 @@ jobs:
     runs-on: macos-latest
     steps:
       - name: Checkout code
-        uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2
+        uses: actions/checkout@9c091bb21b7c1c1d1991bb908d89e4e9dddfe3e0 # v7.0.0
         with:
           persist-credentials: false
 
       - name: Install Go
-        uses: actions/setup-go@4b73464bb391d4059bd26b0524d20df3927bd417 # v6.3.0
+        uses: actions/setup-go@924ae3a1cded613372ab5595356fb5720e22ba16 # v6.5.0
         with:
           go-version-file: "go.mod"
           cache: false
 
       - name: Cache Go modules
-        uses: actions/cache@27d5ce7f107fe9357f9df03efb73ab90386fccae # v5.0.5
+        uses: actions/cache@2c8a9bd7457de244a408f35966fab2fb45fda9c8 # v6.0.0
         with:
           path: ~/go/pkg/mod
           key: macos-gotest-${{ hashFiles('**/go.sum') }}
@@ -45,10 +45,10 @@ jobs:
         run: git --no-pager diff --exit-code
 
       - name: Test
-        run: NETBIRD_STORE_ENGINE=${{ matrix.store }} CI=true go test -coverprofile=coverage.txt -tags=devcert -exec 'sudo --preserve-env=CI,NETBIRD_STORE_ENGINE' -timeout 5m -p 1 $(go list ./... | grep -v -e /management -e /signal -e /relay -e /proxy -e /combined)
+        run: NETBIRD_STORE_ENGINE=${{ matrix.store }} CI=true go test -coverprofile=coverage.txt -tags 'devcert privileged' -exec 'sudo --preserve-env=CI,NETBIRD_STORE_ENGINE' -timeout 5m -p 1 $(go list ./... | grep -v -e /management -e /signal -e /relay -e /proxy -e /combined -e /client/testutil/privileged)
 
       - name: Upload coverage reports to Codecov
-        uses: codecov/codecov-action@e79a6962e0d4c0c17b229090214935d2e33f8354 #v6.0.1
+        uses: codecov/codecov-action@fb8b3582c8e4def4969c97caa2f19720cb33a72f #v7.0.0
         with:
           token: ${{ secrets.CODECOV_TOKEN }}
           slug: netbirdio/netbird

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

@@ -16,7 +16,7 @@ jobs:
     runs-on: ubuntu-22.04
     steps:
       - name: Checkout code
-        uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2
+        uses: actions/checkout@9c091bb21b7c1c1d1991bb908d89e4e9dddfe3e0 # v7.0.0
         with:
           persist-credentials: false
 
@@ -28,7 +28,7 @@ jobs:
         id: test
         env:
           GO_VERSION: ${{ steps.goversion.outputs.version }}
-        uses: vmactions/freebsd-vm@d1e65811565151536c0c894fff74f06351ed26e6 # v1.4.5
+        uses: vmactions/freebsd-vm@b84ab5559b5a1bb4b8ee2737d2506a16e1737636 # v1.4.8
         with:
           usesh: true
           copyback: false
@@ -48,14 +48,14 @@ jobs:
             export PATH=$PATH:/usr/local/go/bin:$HOME/go/bin
             time go build -o netbird client/main.go
             # check all component except management, since we do not support management server on freebsd
-            time go test -timeout 1m -failfast ./base62/...
+            time go test -tags privileged -timeout 1m -failfast ./base62/...
             # NOTE: without -p1 `client/internal/dns` will fail because of `listen udp4 :33100: bind: address already in use`
-            time go test -timeout 8m -failfast -v -p 1 ./client/...
+            time go test -tags privileged -timeout 8m -failfast -v -p 1 ./client/...
-            time go test -timeout 1m -failfast ./dns/...
+            time go test -tags privileged -timeout 1m -failfast ./dns/...
-            time go test -timeout 1m -failfast ./encryption/...
+            time go test -tags privileged -timeout 1m -failfast ./encryption/...
-            time go test -timeout 1m -failfast ./formatter/...
+            time go test -tags privileged -timeout 1m -failfast ./formatter/...
-            time go test -timeout 1m -failfast ./client/iface/...
+            time go test -tags privileged -timeout 1m -failfast ./client/iface/...
-            time go test -timeout 1m -failfast ./route/...
+            time go test -tags privileged -timeout 1m -failfast ./route/...
-            time go test -timeout 1m -failfast ./sharedsock/...
+            time go test -tags privileged -timeout 1m -failfast ./sharedsock/...
-            time go test -timeout 1m -failfast ./util/...
+            time go test -tags privileged -timeout 1m -failfast ./util/...
-            time go test -timeout 1m -failfast ./version/...
+            time go test -tags privileged -timeout 1m -failfast ./version/...

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

@@ -18,7 +18,7 @@ jobs:
       management: ${{ steps.filter.outputs.management }}
     steps:
       - name: Checkout code
-        uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2
+        uses: actions/checkout@9c091bb21b7c1c1d1991bb908d89e4e9dddfe3e0 # v7.0.0
         with:
           persist-credentials: false
 
@@ -30,7 +30,7 @@ jobs:
               - 'management/**'
 
       - name: Install Go
-        uses: actions/setup-go@4b73464bb391d4059bd26b0524d20df3927bd417 # v6.3.0
+        uses: actions/setup-go@924ae3a1cded613372ab5595356fb5720e22ba16 # v6.5.0
         with:
           go-version-file: "go.mod"
           cache: false
@@ -41,7 +41,7 @@ jobs:
           echo "modcache=$(go env GOMODCACHE)" >> $GITHUB_ENV
 
       - name: Cache Go modules
-        uses: actions/cache@27d5ce7f107fe9357f9df03efb73ab90386fccae # v5.0.5
+        uses: actions/cache@2c8a9bd7457de244a408f35966fab2fb45fda9c8 # v6.0.0
         id: cache
         with:
           path: |
@@ -119,12 +119,12 @@ jobs:
     runs-on: ubuntu-22.04
     steps:
       - name: Checkout code
-        uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2
+        uses: actions/checkout@9c091bb21b7c1c1d1991bb908d89e4e9dddfe3e0 # v7.0.0
         with:
           persist-credentials: false
 
       - name: Install Go
-        uses: actions/setup-go@4b73464bb391d4059bd26b0524d20df3927bd417 # v6.3.0
+        uses: actions/setup-go@924ae3a1cded613372ab5595356fb5720e22ba16 # v6.5.0
         with:
           go-version-file: "go.mod"
           cache: false
@@ -135,7 +135,7 @@ jobs:
           echo "modcache=$(go env GOMODCACHE)" >> $GITHUB_ENV
 
       - name: Cache Go modules
-        uses: actions/cache/restore@27d5ce7f107fe9357f9df03efb73ab90386fccae # v5.0.5
+        uses: actions/cache/restore@2c8a9bd7457de244a408f35966fab2fb45fda9c8 # v6.0.0
         with:
           path: |
             ${{ env.cache }}
@@ -158,11 +158,11 @@ jobs:
         run: git --no-pager diff --exit-code
 
       - name: Test
-        run: CGO_ENABLED=1 GOARCH=${{ matrix.arch }} CI=true go test -coverprofile=coverage.txt -tags devcert -exec 'sudo' -timeout 10m -p 1 $(go list ./... | grep -v -e /management -e /signal -e /relay -e /proxy -e /combined)
+        run: CGO_ENABLED=1 GOARCH=${{ matrix.arch }} CI=true go test -coverprofile=coverage.txt -tags devcert -timeout 10m -p 1 $(go list ./... | grep -v -e /management -e /signal -e /relay -e /proxy -e /combined)
 
       - name: Upload coverage reports to Codecov
         if: matrix.arch == 'amd64'
-        uses: codecov/codecov-action@e79a6962e0d4c0c17b229090214935d2e33f8354 #v6.0.1
+        uses: codecov/codecov-action@fb8b3582c8e4def4969c97caa2f19720cb33a72f #v7.0.0
         with:
           token: ${{ secrets.CODECOV_TOKEN }}
           slug: netbirdio/netbird
@@ -175,12 +175,12 @@ jobs:
     runs-on: ubuntu-22.04
     steps:
       - name: Checkout code
-        uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2
+        uses: actions/checkout@9c091bb21b7c1c1d1991bb908d89e4e9dddfe3e0 # v7.0.0
         with:
           persist-credentials: false
 
       - name: Install Go
-        uses: actions/setup-go@4b73464bb391d4059bd26b0524d20df3927bd417 # v6.3.0
+        uses: actions/setup-go@924ae3a1cded613372ab5595356fb5720e22ba16 # v6.5.0
         with:
           go-version-file: "go.mod"
           cache: false
@@ -192,7 +192,7 @@ jobs:
           echo "modcache_dir=$(go env GOMODCACHE)" >> $GITHUB_OUTPUT
 
       - name: Cache Go modules
-        uses: actions/cache/restore@27d5ce7f107fe9357f9df03efb73ab90386fccae # v5.0.5
+        uses: actions/cache/restore@2c8a9bd7457de244a408f35966fab2fb45fda9c8 # v6.0.0
         id: cache-restore
         with:
           path: |
@@ -229,7 +229,7 @@ jobs:
             sh -c ' \
               apk update; apk add --no-cache \
                 ca-certificates iptables ip6tables dbus dbus-dev libpcap-dev build-base; \
-              go test -buildvcs=false -tags devcert -v -timeout 10m -p 1 $(go list -buildvcs=false ./... | grep -v -e /management -e /signal -e /relay -e /proxy -e /combined -e /client/ui -e /upload-server)
+              go test -buildvcs=false -tags "devcert privileged" -v -timeout 10m -p 1 $(go list -buildvcs=false ./... | grep -v -e /management -e /signal -e /relay -e /proxy -e /combined -e /client/ui -e /upload-server -e /client/testutil/privileged)
             '
 
   test_relay:
@@ -246,12 +246,12 @@ jobs:
     runs-on: ubuntu-22.04
     steps:
       - name: Checkout code
-        uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2
+        uses: actions/checkout@9c091bb21b7c1c1d1991bb908d89e4e9dddfe3e0 # v7.0.0
         with:
           persist-credentials: false
 
       - name: Install Go
-        uses: actions/setup-go@4b73464bb391d4059bd26b0524d20df3927bd417 # v6.3.0
+        uses: actions/setup-go@924ae3a1cded613372ab5595356fb5720e22ba16 # v6.5.0
         with:
           go-version-file: "go.mod"
           cache: false
@@ -266,7 +266,7 @@ jobs:
           echo "modcache=$(go env GOMODCACHE)" >> $GITHUB_ENV
 
       - name: Cache Go modules
-        uses: actions/cache/restore@27d5ce7f107fe9357f9df03efb73ab90386fccae # v5.0.5
+        uses: actions/cache/restore@2c8a9bd7457de244a408f35966fab2fb45fda9c8 # v6.0.0
         with:
           path: |
             ${{ env.cache }}
@@ -290,7 +290,7 @@ jobs:
 
       - name: Upload coverage reports to Codecov
         if: matrix.arch == 'amd64'
-        uses: codecov/codecov-action@e79a6962e0d4c0c17b229090214935d2e33f8354 #v6.0.1
+        uses: codecov/codecov-action@fb8b3582c8e4def4969c97caa2f19720cb33a72f #v7.0.0
         with:
           token: ${{ secrets.CODECOV_TOKEN }}
           slug: netbirdio/netbird
@@ -306,12 +306,12 @@ jobs:
     runs-on: ubuntu-22.04
     steps:
       - name: Checkout code
-        uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2
+        uses: actions/checkout@9c091bb21b7c1c1d1991bb908d89e4e9dddfe3e0 # v7.0.0
         with:
           persist-credentials: false
 
       - name: Install Go
-        uses: actions/setup-go@4b73464bb391d4059bd26b0524d20df3927bd417 # v6.3.0
+        uses: actions/setup-go@924ae3a1cded613372ab5595356fb5720e22ba16 # v6.5.0
         with:
           go-version-file: "go.mod"
           cache: false
@@ -325,7 +325,7 @@ jobs:
           echo "modcache=$(go env GOMODCACHE)" >> $GITHUB_ENV
 
       - name: Cache Go modules
-        uses: actions/cache/restore@27d5ce7f107fe9357f9df03efb73ab90386fccae # v5.0.5
+        uses: actions/cache/restore@2c8a9bd7457de244a408f35966fab2fb45fda9c8 # v6.0.0
         with:
           path: |
             ${{ env.cache }}
@@ -347,7 +347,7 @@ jobs:
 
       - name: Upload coverage reports to Codecov
         if: matrix.arch == 'amd64'
-        uses: codecov/codecov-action@e79a6962e0d4c0c17b229090214935d2e33f8354 #v6.0.1
+        uses: codecov/codecov-action@fb8b3582c8e4def4969c97caa2f19720cb33a72f #v7.0.0
         with:
           token: ${{ secrets.CODECOV_TOKEN }}
           slug: netbirdio/netbird
@@ -363,12 +363,12 @@ jobs:
     runs-on: ubuntu-22.04
     steps:
       - name: Checkout code
-        uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2
+        uses: actions/checkout@9c091bb21b7c1c1d1991bb908d89e4e9dddfe3e0 # v7.0.0
         with:
           persist-credentials: false
 
       - name: Install Go
-        uses: actions/setup-go@4b73464bb391d4059bd26b0524d20df3927bd417 # v6.3.0
+        uses: actions/setup-go@924ae3a1cded613372ab5595356fb5720e22ba16 # v6.5.0
         with:
           go-version-file: "go.mod"
           cache: false
@@ -383,7 +383,7 @@ jobs:
           echo "modcache=$(go env GOMODCACHE)" >> $GITHUB_ENV
 
       - name: Cache Go modules
-        uses: actions/cache/restore@27d5ce7f107fe9357f9df03efb73ab90386fccae # v5.0.5
+        uses: actions/cache/restore@2c8a9bd7457de244a408f35966fab2fb45fda9c8 # v6.0.0
         with:
           path: |
             ${{ env.cache }}
@@ -407,7 +407,7 @@ jobs:
 
       - name: Upload coverage reports to Codecov
         if: matrix.arch == 'amd64'
-        uses: codecov/codecov-action@e79a6962e0d4c0c17b229090214935d2e33f8354 #v6.0.1
+        uses: codecov/codecov-action@fb8b3582c8e4def4969c97caa2f19720cb33a72f #v7.0.0
         with:
           token: ${{ secrets.CODECOV_TOKEN }}
           slug: netbirdio/netbird
@@ -424,12 +424,12 @@ jobs:
     runs-on: ubuntu-22.04
     steps:
       - name: Checkout code
-        uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2
+        uses: actions/checkout@9c091bb21b7c1c1d1991bb908d89e4e9dddfe3e0 # v7.0.0
         with:
           persist-credentials: false
 
       - name: Install Go
-        uses: actions/setup-go@4b73464bb391d4059bd26b0524d20df3927bd417 # v6.3.0
+        uses: actions/setup-go@924ae3a1cded613372ab5595356fb5720e22ba16 # v6.5.0
         with:
           go-version-file: "go.mod"
           cache: false
@@ -440,7 +440,7 @@ jobs:
           echo "modcache=$(go env GOMODCACHE)" >> $GITHUB_ENV
 
       - name: Cache Go modules
-        uses: actions/cache/restore@27d5ce7f107fe9357f9df03efb73ab90386fccae # v5.0.5
+        uses: actions/cache/restore@2c8a9bd7457de244a408f35966fab2fb45fda9c8 # v6.0.0
         with:
           path: |
             ${{ env.cache }}
@@ -484,7 +484,7 @@ jobs:
 
       - name: Upload coverage reports to Codecov
         if: matrix.arch == 'amd64'
-        uses: codecov/codecov-action@e79a6962e0d4c0c17b229090214935d2e33f8354 #v6.0.1
+        uses: codecov/codecov-action@fb8b3582c8e4def4969c97caa2f19720cb33a72f #v7.0.0
         with:
           token: ${{ secrets.CODECOV_TOKEN }}
           slug: netbirdio/netbird
@@ -529,12 +529,12 @@ jobs:
             prom/prometheus
 
       - name: Checkout code
-        uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2
+        uses: actions/checkout@9c091bb21b7c1c1d1991bb908d89e4e9dddfe3e0 # v7.0.0
         with:
           persist-credentials: false
 
       - name: Install Go
-        uses: actions/setup-go@4b73464bb391d4059bd26b0524d20df3927bd417 # v6.3.0
+        uses: actions/setup-go@924ae3a1cded613372ab5595356fb5720e22ba16 # v6.5.0
         with:
           go-version-file: "go.mod"
           cache: false
@@ -545,7 +545,7 @@ jobs:
           echo "modcache=$(go env GOMODCACHE)" >> $GITHUB_ENV
 
       - name: Cache Go modules
-        uses: actions/cache/restore@27d5ce7f107fe9357f9df03efb73ab90386fccae # v5.0.5
+        uses: actions/cache/restore@2c8a9bd7457de244a408f35966fab2fb45fda9c8 # v6.0.0
         with:
           path: |
             ${{ env.cache }}
@@ -579,10 +579,11 @@ jobs:
           CGO_ENABLED=1 GOARCH=${{ matrix.arch }} \
           NETBIRD_STORE_ENGINE=${{ matrix.store }} \
           CI=true \
-          GIT_BRANCH=${{ github.ref_name }} \
           go test -tags devcert -run=^$ -bench=. \
           -exec 'sudo --preserve-env=CI,NETBIRD_STORE_ENGINE,GIT_BRANCH,GITHUB_RUN_ID' \
           -timeout 20m ./management/... ./shared/management/... $(go list ./management/... ./shared/management/... | grep -v -e /management/server/http)
+        env:
+          GIT_BRANCH: ${{ github.ref_name }}
 
   api_benchmark:
     name: "Management / Benchmark (API)"
@@ -623,12 +624,12 @@ jobs:
             prom/prometheus
 
       - name: Checkout code
-        uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2
+        uses: actions/checkout@9c091bb21b7c1c1d1991bb908d89e4e9dddfe3e0 # v7.0.0
         with:
           persist-credentials: false
 
       - name: Install Go
-        uses: actions/setup-go@4b73464bb391d4059bd26b0524d20df3927bd417 # v6.3.0
+        uses: actions/setup-go@924ae3a1cded613372ab5595356fb5720e22ba16 # v6.5.0
         with:
           go-version-file: "go.mod"
           cache: false
@@ -639,7 +640,7 @@ jobs:
           echo "modcache=$(go env GOMODCACHE)" >> $GITHUB_ENV
 
       - name: Cache Go modules
-        uses: actions/cache/restore@27d5ce7f107fe9357f9df03efb73ab90386fccae # v5.0.5
+        uses: actions/cache/restore@2c8a9bd7457de244a408f35966fab2fb45fda9c8 # v6.0.0
         with:
           path: |
             ${{ env.cache }}
@@ -673,12 +674,13 @@ jobs:
           CGO_ENABLED=1 GOARCH=${{ matrix.arch }} \
           NETBIRD_STORE_ENGINE=${{ matrix.store }} \
           CI=true \
-          GIT_BRANCH=${{ github.ref_name }} \
           go test -tags=benchmark \
             -run=^$ \
             -bench=. \
             -exec 'sudo --preserve-env=CI,NETBIRD_STORE_ENGINE,GIT_BRANCH,GITHUB_RUN_ID' \
             -timeout 20m ./management/server/http/...
+        env:
+          GIT_BRANCH: ${{ github.ref_name }}
 
   api_integration_test:
     name: "Management / Integration"
@@ -692,12 +694,12 @@ jobs:
     runs-on: ubuntu-22.04
     steps:
       - name: Checkout code
-        uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2
+        uses: actions/checkout@9c091bb21b7c1c1d1991bb908d89e4e9dddfe3e0 # v7.0.0
         with:
           persist-credentials: false
 
       - name: Install Go
-        uses: actions/setup-go@4b73464bb391d4059bd26b0524d20df3927bd417 # v6.3.0
+        uses: actions/setup-go@924ae3a1cded613372ab5595356fb5720e22ba16 # v6.5.0
         with:
           go-version-file: "go.mod"
           cache: false
@@ -708,7 +710,7 @@ jobs:
           echo "modcache=$(go env GOMODCACHE)" >> $GITHUB_ENV
 
       - name: Cache Go modules
-        uses: actions/cache/restore@27d5ce7f107fe9357f9df03efb73ab90386fccae # v5.0.5
+        uses: actions/cache/restore@2c8a9bd7457de244a408f35966fab2fb45fda9c8 # v6.0.0
         with:
           path: |
             ${{ env.cache }}
@@ -734,7 +736,7 @@ jobs:
 
       - name: Upload coverage reports to Codecov
         if: matrix.arch == 'amd64'
-        uses: codecov/codecov-action@e79a6962e0d4c0c17b229090214935d2e33f8354 #v6.0.1
+        uses: codecov/codecov-action@fb8b3582c8e4def4969c97caa2f19720cb33a72f #v7.0.0
         with:
           token: ${{ secrets.CODECOV_TOKEN }}
           slug: netbirdio/netbird

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

@@ -18,12 +18,12 @@ jobs:
     runs-on: windows-latest
     steps:
       - name: Checkout code
-        uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2
+        uses: actions/checkout@9c091bb21b7c1c1d1991bb908d89e4e9dddfe3e0 # v7.0.0
         with:
           persist-credentials: false
 
       - name: Install Go
-        uses: actions/setup-go@4b73464bb391d4059bd26b0524d20df3927bd417 # v6.3.0
+        uses: actions/setup-go@924ae3a1cded613372ab5595356fb5720e22ba16 # v6.5.0
         id: go
         with:
           go-version-file: "go.mod"
@@ -35,7 +35,7 @@ jobs:
           echo "modcache=$(go env GOMODCACHE)" >> $env:GITHUB_ENV
 
       - name: Cache Go modules
-        uses: actions/cache@27d5ce7f107fe9357f9df03efb73ab90386fccae # v5.0.5
+        uses: actions/cache@2c8a9bd7457de244a408f35966fab2fb45fda9c8 # v6.0.0
         with:
           path: |
             ${{ env.cache }}
@@ -68,7 +68,7 @@ jobs:
         run: |
           $packages = go list ./... | Where-Object { $_ -notmatch '/management' } | Where-Object { $_ -notmatch '/relay' } | Where-Object { $_ -notmatch '/signal' } | Where-Object { $_ -notmatch '/proxy' } | Where-Object { $_ -notmatch '/combined' }
           $goExe = "C:\hostedtoolcache\windows\go\${{ steps.go.outputs.go-version }}\x64\bin\go.exe"
-          $cmd = "$goExe test -tags=devcert -timeout 10m -p 1 $($packages -join ' ') > test-out.txt 2>&1"
+          $cmd = "$goExe test -tags `"devcert privileged`" -timeout 10m -p 1 $($packages -join ' ') > test-out.txt 2>&1"
           Set-Content -Path "${{ github.workspace }}\run-tests.cmd" -Value $cmd
 
       - name: test

.github/workflows/golangci-lint.yml

@@ -15,7 +15,7 @@ jobs:
     runs-on: ubuntu-latest
     steps:
       - name: Checkout code
-        uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2
+        uses: actions/checkout@9c091bb21b7c1c1d1991bb908d89e4e9dddfe3e0 # v7.0.0
         with:
           persist-credentials: false
       - name: codespell
@@ -40,7 +40,7 @@ jobs:
     timeout-minutes: 15
     steps:
       - name: Checkout code
-        uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2
+        uses: actions/checkout@9c091bb21b7c1c1d1991bb908d89e4e9dddfe3e0 # v7.0.0
         with:
           persist-credentials: false
       - name: Check for duplicate constants
@@ -48,7 +48,7 @@ jobs:
         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@4b73464bb391d4059bd26b0524d20df3927bd417 # v6.3.0
+        uses: actions/setup-go@924ae3a1cded613372ab5595356fb5720e22ba16 # v6.5.0
         with:
           go-version-file: "go.mod"
           cache: false

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

@@ -22,7 +22,7 @@ jobs:
     runs-on: ${{ matrix.os }}
     steps:
       - name: Checkout code
-        uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2
+        uses: actions/checkout@9c091bb21b7c1c1d1991bb908d89e4e9dddfe3e0 # v7.0.0
         with:
           persist-credentials: false
 

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

@@ -16,11 +16,11 @@ jobs:
     runs-on: ubuntu-latest
     steps:
       - name: Checkout repository
-        uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2
+        uses: actions/checkout@9c091bb21b7c1c1d1991bb908d89e4e9dddfe3e0 # v7.0.0
         with:
           persist-credentials: false
       - name: Install Go
-        uses: actions/setup-go@4b73464bb391d4059bd26b0524d20df3927bd417 # v6.3.0
+        uses: actions/setup-go@924ae3a1cded613372ab5595356fb5720e22ba16 # v6.5.0
         with:
           go-version-file: "go.mod"
       - name: Setup Android SDK
@@ -28,13 +28,13 @@ jobs:
         with:
           cmdline-tools-version: 8512546
       - name: Setup Java
-        uses: actions/setup-java@be666c2fcd27ec809703dec50e508c2fdc7f6654
+        uses: actions/setup-java@1bcf9fb12cf4aa7d266a90ae39939e61372fe520
         with:
           java-version: "11"
           distribution: "adopt"
       - name: NDK Cache
         id: ndk-cache
-        uses: actions/cache@27d5ce7f107fe9357f9df03efb73ab90386fccae # v5.0.5
+        uses: actions/cache@2c8a9bd7457de244a408f35966fab2fb45fda9c8 # v6.0.0
         with:
           path: /usr/local/lib/android/sdk/ndk
           key: ndk-cache-23.1.7779620
@@ -54,11 +54,11 @@ jobs:
     runs-on: macos-latest
     steps:
       - name: Checkout repository
-        uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2
+        uses: actions/checkout@9c091bb21b7c1c1d1991bb908d89e4e9dddfe3e0 # v7.0.0
         with:
           persist-credentials: false
       - name: Install Go
-        uses: actions/setup-go@4b73464bb391d4059bd26b0524d20df3927bd417 # v6.3.0
+        uses: actions/setup-go@924ae3a1cded613372ab5595356fb5720e22ba16 # v6.5.0
         with:
           go-version-file: "go.mod"
       - name: install gomobile

.github/workflows/release.yml

@@ -9,10 +9,13 @@ on:
   pull_request:
 
 env:
-  SIGN_PIPE_VER: "v0.1.5"
+  SIGN_PIPE_VER: "v0.1.6"
-  GORELEASER_VER: "v2.14.3"
+  GORELEASER_VER: "v2.16.0"
   PRODUCT_NAME: "NetBird"
   COPYRIGHT: "NetBird GmbH"
+  flags: ""
+  SKIP_PUBLISH: "true"
+  SKIP_DOCKER_PUSH: "false"
 
 concurrency:
   group: ${{ github.workflow }}-${{ github.ref }}-${{ github.head_ref || github.actor_id }}
@@ -24,7 +27,7 @@ jobs:
     runs-on: ubuntu-22.04
     steps:
       - name: Checkout
-        uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2
+        uses: actions/checkout@9c091bb21b7c1c1d1991bb908d89e4e9dddfe3e0 # v7.0.0
         with:
           persist-credentials: false
 
@@ -61,7 +64,7 @@ jobs:
         if: steps.check_diff.outputs.diff_exists == 'true'
         env:
           GO_VERSION: ${{ steps.goversion.outputs.version }}
-        uses: vmactions/freebsd-vm@d1e65811565151536c0c894fff74f06351ed26e6 # v1.4.5
+        uses: vmactions/freebsd-vm@b84ab5559b5a1bb4b8ee2737d2506a16e1737636 # v1.4.8
         with:
           usesh: true
           copyback: false
@@ -130,11 +133,9 @@ jobs:
       windows_packages_artifact_url: ${{ steps.upload_windows_packages.outputs.artifact-url }}
       macos_packages_artifact_url: ${{ steps.upload_macos_packages.outputs.artifact-url }}
       ghcr_images: ${{ steps.tag_and_push_images.outputs.images_markdown }}
-    env:
-      flags: ""
     steps:
       - name: Checkout
-        uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2
+        uses: actions/checkout@9c091bb21b7c1c1d1991bb908d89e4e9dddfe3e0 # v7.0.0
         with:
           fetch-depth: 0 # It is required for GoReleaser to work properly
           persist-credentials: false
@@ -143,15 +144,34 @@ jobs:
         id: semver_parser
         uses: netbirdio/shared-actions/actions/parse-semver@be5df6047383da2236e02243cceb857d8567c27e # v0.0.2
 
-      - if: ${{ !startsWith(github.ref, 'refs/tags/v') }}
+      - name: Set snapshot flag
-        run: echo "flags=--snapshot" >> $GITHUB_ENV
+        if: ${{ !startsWith(github.ref, 'refs/tags/v') }}
+        run: | 
+          echo "flags=--snapshot" >> $GITHUB_ENV
+
+      - name: Set build vars
+        if: ${{ startsWith(github.ref, 'refs/tags/v') }}
+        run: |
+          if [[ "x-${{ steps.semver_parser.outputs.prerelease }}" == "x-" && "x-${{ github.repository }}" == "x-netbirdio/netbird" ]]; then
+            echo "x-${{ github.repository }}" 
+            echo "x-${{ steps.semver_parser.outputs.prerelease }}"
+            echo "SKIP_PUBLISH=false" >> $GITHUB_ENV
+          else
+            echo "x-${{ github.repository }}" 
+            echo "x-${{ steps.semver_parser.outputs.prerelease }}"
+          fi
+          
+          if [[ "x-${{ github.repository }}" != "x-netbirdio/netbird" ]]; then
+            echo "SKIP_DOCKER_PUSH=true" >> $GITHUB_ENV
+          fi
+
       - name: Set up Go
-        uses: actions/setup-go@4b73464bb391d4059bd26b0524d20df3927bd417 # v6.3.0
+        uses: actions/setup-go@924ae3a1cded613372ab5595356fb5720e22ba16 # v6.5.0
         with:
           go-version-file: "go.mod"
           cache: false
       - name: Cache Go modules
-        uses: actions/cache/restore@27d5ce7f107fe9357f9df03efb73ab90386fccae # v5.0.5
+        uses: actions/cache/restore@2c8a9bd7457de244a408f35966fab2fb45fda9c8 # v6.0.0
         with:
           path: |
             ~/go/pkg/mod
@@ -161,12 +181,14 @@ jobs:
             ${{ runner.os }}-go-releaser-
       - name: Install modules
         run: go mod tidy
+      - name: run openapi generator
+        run: bash shared/management/http/api/generate.sh
       - name: check git status
         run: git --no-pager diff --exit-code
       - name: Set up QEMU
-        uses: docker/setup-qemu-action@ce360397dd3f832beb865e1373c09c0e9f86d70a #v4.0.0
+        uses: docker/setup-qemu-action@06116385d9baf250c9f4dcb4858b16962ea869c3 #v4.1.0
       - name: Set up Docker Buildx
-        uses: docker/setup-buildx-action@4d04d5d9486b7bd6fa91e7baf45bbb4f8b9deedd #v4.0.0
+        uses: docker/setup-buildx-action@d7f5e7f509e45cec5c76c4d5afdd7de93d0b3df5 #v4.1.0
       - name: Login to Docker hub
         if: github.event_name != 'pull_request'
         uses: docker/login-action@650006c6eb7dba73a995cc03b0b2d7f5ca915bee # v4.2.0
@@ -199,7 +221,7 @@ jobs:
         run: goversioninfo -arm -64 -icon client/ui/assets/netbird.ico -manifest client/manifest.xml -product-name ${{ env.PRODUCT_NAME }} -copyright "${{ env.COPYRIGHT }}" -ver-major ${{ steps.semver_parser.outputs.major }} -ver-minor ${{ steps.semver_parser.outputs.minor }} -ver-patch ${{ steps.semver_parser.outputs.patch }} -ver-build 0 -file-version ${{ steps.semver_parser.outputs.fullversion }}.0 -product-version ${{ steps.semver_parser.outputs.fullversion }}.0 -o client/resources_windows_arm64.syso
       - name: Run GoReleaser
         id: goreleaser
-        uses: goreleaser/goreleaser-action@4c6ab561adb47e50c45ef534e2155934e91c40c1 # v7.2.0
+        uses: goreleaser/goreleaser-action@5daf1e915a5f0af01ddbcd89a43b8061ff4f1a89 # v7.2.2
         with:
           version: ${{ env.GORELEASER_VER }}
           args: release --clean ${{ env.flags }}
@@ -210,6 +232,8 @@ jobs:
           UPLOAD_YUM_SECRET: ${{ secrets.PKG_UPLOAD_SECRET }}
           GPG_RPM_KEY_FILE: ${{ env.GPG_RPM_KEY_FILE }}
           NFPM_NETBIRD_RPM_PASSPHRASE: ${{ secrets.GPG_RPM_PASSPHRASE }}
+          SKIP_PUBLISH: ${{ env.SKIP_PUBLISH }}
+          SKIP_DOCKER_PUSH: ${{ env.SKIP_DOCKER_PUSH }}
       - name: Verify RPM signatures
         run: |
           docker run --rm -v $(pwd)/dist:/dist fedora:41 bash -c '
@@ -323,7 +347,7 @@ jobs:
       release_ui_artifact_url: ${{ steps.upload_release_ui.outputs.artifact-url }}
     steps:
       - name: Checkout
-        uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2
+        uses: actions/checkout@9c091bb21b7c1c1d1991bb908d89e4e9dddfe3e0 # v7.0.0
         with:
           fetch-depth: 0 # It is required for GoReleaser to work properly
           persist-credentials: false
@@ -332,16 +356,30 @@ jobs:
         id: semver_parser
         uses: netbirdio/shared-actions/actions/parse-semver@be5df6047383da2236e02243cceb857d8567c27e # v0.0.2
 
-      - if: ${{ !startsWith(github.ref, 'refs/tags/v') }}
+      - name: Set snapshot flag
-        run: echo "flags=--snapshot" >> $GITHUB_ENV
+        if: ${{ !startsWith(github.ref, 'refs/tags/v') }}
+        run: |
+          echo "flags=--snapshot" >> $GITHUB_ENV
+
+      - name: Set build vars
+        if: ${{ startsWith(github.ref, 'refs/tags/v') }}
+        run: |
+          if [[ "x-${{ steps.semver_parser.outputs.prerelease }}" == "x-" && "x-${{ github.repository }}" == "x-netbirdio/netbird" ]]; then
+            echo "x-${{ github.repository }}" 
+            echo "x-${{ steps.semver_parser.outputs.prerelease }}"
+            echo "SKIP_PUBLISH=false" >> $GITHUB_ENV
+          else
+            echo "x-${{ github.repository }}" 
+            echo "x-${{ steps.semver_parser.outputs.prerelease }}"
+          fi
 
       - name: Set up Go
-        uses: actions/setup-go@4b73464bb391d4059bd26b0524d20df3927bd417 # v6.3.0
+        uses: actions/setup-go@924ae3a1cded613372ab5595356fb5720e22ba16 # v6.5.0
         with:
           go-version-file: "go.mod"
           cache: false
       - name: Cache Go modules
-        uses: actions/cache@27d5ce7f107fe9357f9df03efb73ab90386fccae # v5.0.5
+        uses: actions/cache@2c8a9bd7457de244a408f35966fab2fb45fda9c8 # v6.0.0
         with:
           path: |
             ~/go/pkg/mod
@@ -382,7 +420,7 @@ jobs:
         run: goversioninfo -arm -64 -icon client/ui/assets/netbird.ico -manifest client/ui/manifest.xml -product-name ${{ env.PRODUCT_NAME }}-"UI" -copyright "${{ env.COPYRIGHT }}" -ver-major ${{ steps.semver_parser.outputs.major }} -ver-minor ${{ steps.semver_parser.outputs.minor }} -ver-patch ${{ steps.semver_parser.outputs.patch }} -ver-build 0 -file-version ${{ steps.semver_parser.outputs.fullversion }}.0 -product-version ${{ steps.semver_parser.outputs.fullversion }}.0 -o client/ui/resources_windows_arm64.syso
 
       - name: Run GoReleaser
-        uses: goreleaser/goreleaser-action@4c6ab561adb47e50c45ef534e2155934e91c40c1 # v7.2.0
+        uses: goreleaser/goreleaser-action@5daf1e915a5f0af01ddbcd89a43b8061ff4f1a89 # v7.2.2
         with:
           version: ${{ env.GORELEASER_VER }}
           args: release --config .goreleaser_ui.yaml --clean ${{ env.flags }}
@@ -393,6 +431,7 @@ jobs:
           UPLOAD_YUM_SECRET: ${{ secrets.PKG_UPLOAD_SECRET }}
           GPG_RPM_KEY_FILE: ${{ env.GPG_RPM_KEY_FILE }}
           NFPM_NETBIRD_UI_RPM_PASSPHRASE: ${{ secrets.GPG_RPM_PASSPHRASE }}
+          SKIP_PUBLISH: ${{ env.SKIP_PUBLISH }}
       - name: Verify RPM signatures
         run: |
           docker run --rm -v $(pwd)/dist:/dist fedora:41 bash -c '
@@ -425,17 +464,17 @@ jobs:
       - if: ${{ !startsWith(github.ref, 'refs/tags/v') }}
         run: echo "flags=--snapshot" >> $GITHUB_ENV
       - name: Checkout
-        uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2
+        uses: actions/checkout@9c091bb21b7c1c1d1991bb908d89e4e9dddfe3e0 # v7.0.0
         with:
           fetch-depth: 0 # It is required for GoReleaser to work properly
           persist-credentials: false
       - name: Set up Go
-        uses: actions/setup-go@4b73464bb391d4059bd26b0524d20df3927bd417 # v6.3.0
+        uses: actions/setup-go@924ae3a1cded613372ab5595356fb5720e22ba16 # v6.5.0
         with:
           go-version-file: "go.mod"
           cache: false
       - name: Cache Go modules
-        uses: actions/cache@27d5ce7f107fe9357f9df03efb73ab90386fccae # v5.0.5
+        uses: actions/cache@2c8a9bd7457de244a408f35966fab2fb45fda9c8 # v6.0.0
         with:
           path: |
             ~/go/pkg/mod
@@ -449,7 +488,7 @@ jobs:
         run: git --no-pager diff --exit-code
       - name: Run GoReleaser
         id: goreleaser
-        uses: goreleaser/goreleaser-action@4c6ab561adb47e50c45ef534e2155934e91c40c1 # v7.2.0
+        uses: goreleaser/goreleaser-action@5daf1e915a5f0af01ddbcd89a43b8061ff4f1a89 # v7.2.2
         with:
           version: ${{ env.GORELEASER_VER }}
           args: release --config .goreleaser_ui_darwin.yaml --clean ${{ env.flags }}
@@ -483,7 +522,7 @@ jobs:
       downloadPath: '${{ github.workspace }}\temp'
     steps:
       - name: Checkout
-        uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2
+        uses: actions/checkout@9c091bb21b7c1c1d1991bb908d89e4e9dddfe3e0 # v7.0.0
         with:
           persist-credentials: false
 
@@ -495,13 +534,13 @@ jobs:
         run: echo "C:\Program Files\7-Zip" | Out-File -FilePath $env:GITHUB_PATH -Encoding utf8 -Append
 
       - name: Download release artifacts
-        uses: actions/download-artifact@70fc10c6e5e1ce46ad2ea6f2b72d43f7d47b13c3 # v8.0.1
+        uses: actions/download-artifact@3e5f45b2cfb9172054b4087a40e8e0b5a5461e7c # v8.0.1
         with:
           name: release
           path: release
 
       - name: Download UI release artifacts
-        uses: actions/download-artifact@70fc10c6e5e1ce46ad2ea6f2b72d43f7d47b13c3 # v8.0.1
+        uses: actions/download-artifact@3e5f45b2cfb9172054b4087a40e8e0b5a5461e7c # v8.0.1
         with:
           name: release-ui
           path: release-ui

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

@@ -68,17 +68,17 @@ jobs:
         run: sudo apt-get install -y curl
 
       - name: Checkout code
-        uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2
+        uses: actions/checkout@9c091bb21b7c1c1d1991bb908d89e4e9dddfe3e0 # v7.0.0
         with:
           persist-credentials: false
 
       - name: Install Go
-        uses: actions/setup-go@4b73464bb391d4059bd26b0524d20df3927bd417 # v6.3.0
+        uses: actions/setup-go@924ae3a1cded613372ab5595356fb5720e22ba16 # v6.5.0
         with:
           go-version-file: "go.mod"
 
       - name: Cache Go modules
-        uses: actions/cache@27d5ce7f107fe9357f9df03efb73ab90386fccae # v5.0.5
+        uses: actions/cache@2c8a9bd7457de244a408f35966fab2fb45fda9c8 # v6.0.0
         with:
           path: ~/go/pkg/mod
           key: ${{ runner.os }}-go-${{ hashFiles('**/go.sum') }}
@@ -207,7 +207,7 @@ jobs:
       - name: Build management docker image
         working-directory: management
         run: |
-          docker build -t netbirdio/management:latest .
+          docker build -t netbirdio/management:latest --build-arg TARGETPLATFORM=. .
 
       - name: Build signal binary
         working-directory: signal
@@ -216,7 +216,7 @@ jobs:
       - name: Build signal docker image
         working-directory: signal
         run: |
-          docker build -t netbirdio/signal:latest .
+          docker build -t netbirdio/signal:latest --build-arg TARGETPLATFORM=. .
 
       - name: Build relay binary
         working-directory: relay
@@ -225,7 +225,7 @@ jobs:
       - name: Build relay docker image
         working-directory: relay
         run: |
-          docker build -t netbirdio/relay:latest .
+          docker build -t netbirdio/relay:latest --build-arg TARGETPLATFORM=. .
 
       - name: run docker compose up
         working-directory: infrastructure_files/artifacts
@@ -256,7 +256,7 @@ jobs:
         run: sudo apt-get install -y jq
 
       - name: Checkout code
-        uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2
+        uses: actions/checkout@9c091bb21b7c1c1d1991bb908d89e4e9dddfe3e0 # v7.0.0
         with:
           persist-credentials: false
 

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

@@ -19,11 +19,11 @@ jobs:
       GOARCH: wasm
     steps:
       - name: Checkout repository
-        uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2
+        uses: actions/checkout@9c091bb21b7c1c1d1991bb908d89e4e9dddfe3e0 # v7.0.0
         with:
           persist-credentials: false
       - name: Install Go
-        uses: actions/setup-go@4b73464bb391d4059bd26b0524d20df3927bd417 # v6.3.0
+        uses: actions/setup-go@924ae3a1cded613372ab5595356fb5720e22ba16 # v6.5.0
         with:
           go-version-file: "go.mod"
       - name: Install dependencies
@@ -44,11 +44,11 @@ jobs:
     runs-on: ubuntu-latest
     steps:
       - name: Checkout repository
-        uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2
+        uses: actions/checkout@9c091bb21b7c1c1d1991bb908d89e4e9dddfe3e0 # v7.0.0
         with:
           persist-credentials: false
       - name: Install Go
-        uses: actions/setup-go@4b73464bb391d4059bd26b0524d20df3927bd417 # v6.3.0
+        uses: actions/setup-go@924ae3a1cded613372ab5595356fb5720e22ba16 # v6.5.0
         with:
           go-version-file: "go.mod"
       - name: Build Wasm client

.goreleaser.yaml

@@ -1,5 +1,7 @@
 version: 2
-
+env:
+  - SKIP_PUBLISH={{ if index .Env "SKIP_PUBLISH" }}{{ .Env.SKIP_PUBLISH }}{{ else }}true{{ end }}
+  - SKIP_DOCKER_PUSH={{ if index .Env "SKIP_DOCKER_PUSH" }}{{ .Env.SKIP_DOCKER_PUSH }}{{ else }}false{{ end }}
 project_name: netbird
 builds:
   - id: netbird-wasm
@@ -74,6 +76,8 @@ builds:
       - amd64
       - arm64
       - arm
+    goarm:
+      - 7
     ldflags:
       - -s -w -X github.com/netbirdio/netbird/version.version={{.Version}} -X main.commit={{.Commit}} -X main.date={{.CommitDate}} -X main.builtBy=goreleaser
     mod_timestamp: "{{ .CommitTimestamp }}"
@@ -88,6 +92,8 @@ builds:
       - amd64
       - arm64
       - arm
+    goarm:
+      - 7
     ldflags:
       - -s -w -X github.com/netbirdio/netbird/version.version={{.Version}} -X main.commit={{.Commit}} -X main.date={{.CommitDate}} -X main.builtBy=goreleaser
     mod_timestamp: "{{ .CommitTimestamp }}"
@@ -102,6 +108,8 @@ builds:
       - amd64
       - arm64
       - arm
+    goarm:
+      - 7
     ldflags:
       - -s -w -X github.com/netbirdio/netbird/version.version={{.Version}} -X main.commit={{.Commit}} -X main.date={{.CommitDate}} -X main.builtBy=goreleaser
     mod_timestamp: "{{ .CommitTimestamp }}"
@@ -122,6 +130,8 @@ builds:
       - amd64
       - arm64
       - arm
+    goarm:
+      - 7
     ldflags:
       - -s -w -X github.com/netbirdio/netbird/version.version={{.Version}} -X main.commit={{.Commit}} -X main.date={{.CommitDate}} -X main.builtBy=goreleaser
     mod_timestamp: "{{ .CommitTimestamp }}"
@@ -136,6 +146,8 @@ builds:
       - amd64
       - arm64
       - arm
+    goarm:
+      - 7
     ldflags:
       - -s -w -X github.com/netbirdio/netbird/version.version={{.Version}} -X main.commit={{.Commit}} -X main.date={{.CommitDate}} -X main.builtBy=goreleaser
     mod_timestamp: "{{ .CommitTimestamp }}"
@@ -150,6 +162,8 @@ builds:
       - amd64
       - arm64
       - arm
+    goarm:
+      - 7
     ldflags:
       - -s -w -X main.Version={{.Version}} -X main.Commit={{.Commit}} -X main.BuildDate={{.CommitDate}}
     mod_timestamp: "{{ .CommitTimestamp }}"
@@ -170,6 +184,8 @@ builds:
       - amd64
       - arm64
       - arm
+    goarm:
+      - 7
     ldflags:
       - -s -w -X github.com/netbirdio/netbird/version.version={{.Version}} -X main.commit={{.Commit}} -X main.date={{.CommitDate}} -X main.builtBy=goreleaser
     mod_timestamp: "{{ .CommitTimestamp }}"
@@ -222,670 +238,192 @@ nfpms:
     rpm:
       signature:
         key_file: '{{ if index .Env "GPG_RPM_KEY_FILE" }}{{ .Env.GPG_RPM_KEY_FILE }}{{ end }}'
-dockers:
+dockers_v2:
-  - image_templates:
+   - id: netbird
-      - netbirdio/netbird:{{ .Version }}-amd64
+     disable: "{{ .Env.SKIP_DOCKER_PUSH }}"
-      - ghcr.io/netbirdio/netbird:{{ .Version }}-amd64
+     ids:
-    ids:
+       - netbird
-      - netbird
+     images:
-    goarch: amd64
+       - netbirdio/netbird
-    use: buildx
+       - ghcr.io/netbirdio/netbird
-    dockerfile: client/Dockerfile
+     tags:
-    extra_files:
+       - "{{ .Version }}"
-      - client/netbird-entrypoint.sh
+       - "{{ if eq .Env.SKIP_PUBLISH \"false\" }}latest{{ end }}"
-    build_flag_templates:
+     dockerfile: client/Dockerfile
-      - "--platform=linux/amd64"
+     extra_files:
-      - "--label=org.opencontainers.image.created={{.Date}}"
+       - client/netbird-entrypoint.sh
-      - "--label=org.opencontainers.image.title={{.ProjectName}}"
+     platforms:
-      - "--label=org.opencontainers.image.version={{.Version}}"
+       - linux/amd64
-      - "--label=org.opencontainers.image.revision={{.FullCommit}}"
+       - linux/arm64
-      - "--label=org.opencontainers.image.source=https://github.com/netbirdio/{{.ProjectName}}"
+       - linux/arm/6
-      - "--label=maintainer=dev@netbird.io"
+     annotations:
-  - image_templates:
+       "org.opencontainers.image.created": "{{.Date}}"
-      - netbirdio/netbird:{{ .Version }}-arm64v8
+       "org.opencontainers.image.title": "{{.ProjectName}}"
-      - ghcr.io/netbirdio/netbird:{{ .Version }}-arm64v8
+       "org.opencontainers.image.version": "{{.Version}}"
-    ids:
+       "org.opencontainers.image.revision": "{{.FullCommit}}"
-      - netbird
+       "org.opencontainers.image.source": "{{.GitURL}}"
-    goarch: arm64
+       "maintainer": "dev@netbird.io"
-    use: buildx
+   - id: netbird-rootless
-    dockerfile: client/Dockerfile
+     disable: "{{ .Env.SKIP_DOCKER_PUSH }}"
-    extra_files:
+     ids:
-      - client/netbird-entrypoint.sh
+       - netbird
-    build_flag_templates:
+     images:
-      - "--platform=linux/arm64"
+       - netbirdio/netbird
-      - "--label=org.opencontainers.image.created={{.Date}}"
+       - ghcr.io/netbirdio/netbird
-      - "--label=org.opencontainers.image.title={{.ProjectName}}"
+     tags:
-      - "--label=org.opencontainers.image.version={{.Version}}"
+       - "v{{ .Version }}-rootless"
-      - "--label=org.opencontainers.image.revision={{.FullCommit}}"
+       - "{{ if eq .Env.SKIP_PUBLISH \"false\" }}latest{{ end }}"
-      - "--label=org.opencontainers.image.source=https://github.com/netbirdio/{{.ProjectName}}"
+     dockerfile: client/Dockerfile-rootless
-      - "--label=maintainer=dev@netbird.io"
+     extra_files:
-  - image_templates:
+       - client/netbird-entrypoint.sh
-      - netbirdio/netbird:{{ .Version }}-arm
+     platforms:
-      - ghcr.io/netbirdio/netbird:{{ .Version }}-arm
+       - linux/amd64
-    ids:
+       - linux/arm64
-      - netbird
+       - linux/arm/6
-    goarch: arm
+     annotations:
-    goarm: 6
+       "org.opencontainers.image.created": "{{.Date}}"
-    use: buildx
+       "org.opencontainers.image.title": "{{.ProjectName}}"
-    dockerfile: client/Dockerfile
+       "org.opencontainers.image.version": "{{.Version}}"
-    extra_files:
+       "org.opencontainers.image.revision": "{{.FullCommit}}"
-      - client/netbird-entrypoint.sh
+       "org.opencontainers.image.source": "{{.GitURL}}"
-    build_flag_templates:
+       "maintainer": "dev@netbird.io"
-      - "--platform=linux/arm"
+   - id: relay
-      - "--label=org.opencontainers.image.created={{.Date}}"
+     disable: "{{ .Env.SKIP_DOCKER_PUSH }}"
-      - "--label=org.opencontainers.image.title={{.ProjectName}}"
+     ids:
-      - "--label=org.opencontainers.image.version={{.Version}}"
+       - netbird-relay
-      - "--label=org.opencontainers.image.revision={{.FullCommit}}"
+     images:
-      - "--label=org.opencontainers.image.source=https://github.com/netbirdio/{{.ProjectName}}"
+       - netbirdio/relay
-      - "--label=maintainer=dev@netbird.io"
+       - ghcr.io/netbirdio/relay
-
+     tags:
-  - image_templates:
+       - "{{ .Version }}"
-      - netbirdio/netbird:{{ .Version }}-rootless-amd64
+       - "{{ if eq .Env.SKIP_PUBLISH \"false\" }}latest{{ end }}"
-      - ghcr.io/netbirdio/netbird:{{ .Version }}-rootless-amd64
+     dockerfile: relay/Dockerfile
-    ids:
+     platforms:
-      - netbird
+       - linux/amd64
-    goarch: amd64
+       - linux/arm64
-    use: buildx
+       - linux/arm
-    dockerfile: client/Dockerfile-rootless
+     annotations:
-    extra_files:
+       "org.opencontainers.image.created": "{{.Date}}"
-      - client/netbird-entrypoint.sh
+       "org.opencontainers.image.title": "{{.ProjectName}}"
-    build_flag_templates:
+       "org.opencontainers.image.version": "{{.Version}}"
-      - "--platform=linux/amd64"
+       "org.opencontainers.image.revision": "{{.FullCommit}}"
-      - "--label=org.opencontainers.image.created={{.Date}}"
+       "org.opencontainers.image.source": "{{.GitURL}}"
-      - "--label=org.opencontainers.image.title={{.ProjectName}}"
+       "maintainer": "dev@netbird.io"
-      - "--label=org.opencontainers.image.version={{.Version}}"
+   - id: signal
-      - "--label=org.opencontainers.image.revision={{.FullCommit}}"
+     disable: "{{ .Env.SKIP_DOCKER_PUSH }}"
-      - "--label=org.opencontainers.image.source=https://github.com/netbirdio/{{.ProjectName}}"
+     ids:
-      - "--label=maintainer=dev@netbird.io"
+       - netbird-signal
-  - image_templates:
+     images:
-      - netbirdio/netbird:{{ .Version }}-rootless-arm64v8
+       - netbirdio/signal
-      - ghcr.io/netbirdio/netbird:{{ .Version }}-rootless-arm64v8
+       - ghcr.io/netbirdio/signal
-    ids:
+     tags:
-      - netbird
+       - "{{ .Version }}"
-    goarch: arm64
+       - "{{ if eq .Env.SKIP_PUBLISH \"false\" }}latest{{ end }}"
-    use: buildx
+     dockerfile: signal/Dockerfile
-    dockerfile: client/Dockerfile-rootless
+     platforms:
-    extra_files:
+       - linux/amd64
-      - client/netbird-entrypoint.sh
+       - linux/arm64
-    build_flag_templates:
+       - linux/arm
-      - "--platform=linux/arm64"
+     annotations:
-      - "--label=org.opencontainers.image.created={{.Date}}"
+       "org.opencontainers.image.created": "{{.Date}}"
-      - "--label=org.opencontainers.image.title={{.ProjectName}}"
+       "org.opencontainers.image.title": "{{.ProjectName}}"
-      - "--label=org.opencontainers.image.version={{.Version}}"
+       "org.opencontainers.image.version": "{{.Version}}"
-      - "--label=org.opencontainers.image.revision={{.FullCommit}}"
+       "org.opencontainers.image.revision": "{{.FullCommit}}"
-      - "--label=org.opencontainers.image.source=https://github.com/netbirdio/{{.ProjectName}}"
+       "org.opencontainers.image.source": "{{.GitURL}}"
-      - "--label=maintainer=dev@netbird.io"
+       "maintainer": "dev@netbird.io"
-  - image_templates:
+   - id: management
-      - netbirdio/netbird:{{ .Version }}-rootless-arm
+     disable: "{{ .Env.SKIP_DOCKER_PUSH }}"
-      - ghcr.io/netbirdio/netbird:{{ .Version }}-rootless-arm
+     ids:
-    ids:
+       - netbird-mgmt
-      - netbird
+     images:
-    goarch: arm
+       - netbirdio/management
-    goarm: 6
+       - ghcr.io/netbirdio/management
-    use: buildx
+     tags:
-    dockerfile: client/Dockerfile-rootless
+       - "{{ .Version }}"
-    extra_files:
+       - "{{ if eq .Env.SKIP_PUBLISH \"false\" }}latest{{ end }}"
-      - client/netbird-entrypoint.sh
+     dockerfile: management/Dockerfile
-    build_flag_templates:
+     platforms:
-      - "--platform=linux/arm"
+       - linux/amd64
-      - "--label=org.opencontainers.image.created={{.Date}}"
+       - linux/arm64
-      - "--label=org.opencontainers.image.title={{.ProjectName}}"
+       - linux/arm
-      - "--label=org.opencontainers.image.version={{.Version}}"
+     annotations:
-      - "--label=org.opencontainers.image.revision={{.FullCommit}}"
+       "org.opencontainers.image.created": "{{.Date}}"
-      - "--label=org.opencontainers.image.source=https://github.com/netbirdio/{{.ProjectName}}"
+       "org.opencontainers.image.title": "{{.ProjectName}}"
-      - "--label=maintainer=dev@netbird.io"
+       "org.opencontainers.image.version": "{{.Version}}"
-
+       "org.opencontainers.image.revision": "{{.FullCommit}}"
-  - image_templates:
+       "org.opencontainers.image.source": "{{.GitURL}}"
-      - netbirdio/relay:{{ .Version }}-amd64
+       "maintainer": "dev@netbird.io"
-      - ghcr.io/netbirdio/relay:{{ .Version }}-amd64
+   - id: upload
-    ids:
+     disable: "{{ .Env.SKIP_DOCKER_PUSH }}"
-      - netbird-relay
+     ids:
-    goarch: amd64
+       - netbird-upload
-    use: buildx
+     images:
-    dockerfile: relay/Dockerfile
+       - netbirdio/upload
-    build_flag_templates:
+       - ghcr.io/netbirdio/upload
-      - "--platform=linux/amd64"
+     tags:
-      - "--label=org.opencontainers.image.created={{.Date}}"
+       - "{{ .Version }}"
-      - "--label=org.opencontainers.image.title={{.ProjectName}}"
+       - "{{ if eq .Env.SKIP_PUBLISH \"false\" }}latest{{ end }}"
-      - "--label=org.opencontainers.image.version={{.Version}}"
+     dockerfile: upload-server/Dockerfile
-      - "--label=org.opencontainers.image.revision={{.FullCommit}}"
+     platforms:
-      - "--label=org.opencontainers.image.source=https://github.com/netbirdio/{{.ProjectName}}"
+       - linux/amd64
-      - "--label=maintainer=dev@netbird.io"
+       - linux/arm64
-  - image_templates:
+       - linux/arm
-      - netbirdio/relay:{{ .Version }}-arm64v8
+     annotations:
-      - ghcr.io/netbirdio/relay:{{ .Version }}-arm64v8
+       "org.opencontainers.image.created": "{{.Date}}"
-    ids:
+       "org.opencontainers.image.title": "{{.ProjectName}}"
-      - netbird-relay
+       "org.opencontainers.image.version": "{{.Version}}"
-    goarch: arm64
+       "org.opencontainers.image.revision": "{{.FullCommit}}"
-    use: buildx
+       "org.opencontainers.image.source": "{{.GitURL}}"
-    dockerfile: relay/Dockerfile
+       "maintainer": "dev@netbird.io"
-    build_flag_templates:
+   - id: netbird-server
-      - "--platform=linux/arm64"
+     disable: "{{ .Env.SKIP_DOCKER_PUSH }}"
-      - "--label=org.opencontainers.image.created={{.Date}}"
+     ids:
-      - "--label=org.opencontainers.image.title={{.ProjectName}}"
+       - netbird-server
-      - "--label=org.opencontainers.image.version={{.Version}}"
+     images:
-      - "--label=org.opencontainers.image.revision={{.FullCommit}}"
+       - netbirdio/netbird-server
-      - "--label=org.opencontainers.image.source=https://github.com/netbirdio/{{.ProjectName}}"
+       - ghcr.io/netbirdio/netbird-server
-      - "--label=maintainer=dev@netbird.io"
+     tags:
-  - image_templates:
+       - "{{ .Version }}"
-      - netbirdio/relay:{{ .Version }}-arm
+       - "{{ if eq .Env.SKIP_PUBLISH \"false\" }}latest{{ end }}"
-      - ghcr.io/netbirdio/relay:{{ .Version }}-arm
+     dockerfile: combined/Dockerfile
-    ids:
+     platforms:
-      - netbird-relay
+       - linux/amd64
-    goarch: arm
+       - linux/arm64
-    goarm: 6
+       - linux/arm
-    use: buildx
+     annotations:
-    dockerfile: relay/Dockerfile
+       "org.opencontainers.image.created": "{{.Date}}"
-    build_flag_templates:
+       "org.opencontainers.image.title": "{{.ProjectName}}"
-      - "--platform=linux/arm"
+       "org.opencontainers.image.version": "{{.Version}}"
-      - "--label=org.opencontainers.image.created={{.Date}}"
+       "org.opencontainers.image.revision": "{{.FullCommit}}"
-      - "--label=org.opencontainers.image.title={{.ProjectName}}"
+       "org.opencontainers.image.source": "{{.GitURL}}"
-      - "--label=org.opencontainers.image.version={{.Version}}"
+       "maintainer": "dev@netbird.io"
-      - "--label=org.opencontainers.image.revision={{.FullCommit}}"
+   - id: netbird-proxy
-      - "--label=org.opencontainers.image.source=https://github.com/netbirdio/{{.ProjectName}}"
+     disable: "{{ .Env.SKIP_DOCKER_PUSH }}"
-      - "--label=maintainer=dev@netbird.io"
+     ids:
-  - image_templates:
+       - netbird-proxy
-      - netbirdio/signal:{{ .Version }}-amd64
+     images:
-      - ghcr.io/netbirdio/signal:{{ .Version }}-amd64
+       - netbirdio/reverse-proxy
-    ids:
+       - ghcr.io/netbirdio/reverse-proxy
-      - netbird-signal
+     tags:
-    goarch: amd64
+       - "{{ .Version }}"
-    use: buildx
+       - "{{ if eq .Env.SKIP_PUBLISH \"false\" }}latest{{ end }}"
-    dockerfile: signal/Dockerfile
+     dockerfile: proxy/Dockerfile
-    build_flag_templates:
+     platforms:
-      - "--platform=linux/amd64"
+       - linux/amd64
-      - "--label=org.opencontainers.image.created={{.Date}}"
+       - linux/arm64
-      - "--label=org.opencontainers.image.title={{.ProjectName}}"
+       - linux/arm
-      - "--label=org.opencontainers.image.version={{.Version}}"
+     annotations:
-      - "--label=org.opencontainers.image.revision={{.FullCommit}}"
+       "org.opencontainers.image.created": "{{.Date}}"
-      - "--label=org.opencontainers.image.source=https://github.com/netbirdio/{{.ProjectName}}"
+       "org.opencontainers.image.title": "{{.ProjectName}}"
-      - "--label=maintainer=dev@netbird.io"
+       "org.opencontainers.image.version": "{{.Version}}"
-  - image_templates:
+       "org.opencontainers.image.revision": "{{.FullCommit}}"
-      - netbirdio/signal:{{ .Version }}-arm64v8
+       "org.opencontainers.image.source": "{{.GitURL}}"
-      - ghcr.io/netbirdio/signal:{{ .Version }}-arm64v8
+       "maintainer": "dev@netbird.io"
-    ids:
-      - netbird-signal
-    goarch: arm64
-    use: buildx
-    dockerfile: signal/Dockerfile
-    build_flag_templates:
-      - "--platform=linux/arm64"
-      - "--label=org.opencontainers.image.created={{.Date}}"
-      - "--label=org.opencontainers.image.title={{.ProjectName}}"
-      - "--label=org.opencontainers.image.version={{.Version}}"
-      - "--label=org.opencontainers.image.revision={{.FullCommit}}"
-      - "--label=org.opencontainers.image.source=https://github.com/netbirdio/{{.ProjectName}}"
-      - "--label=maintainer=dev@netbird.io"
-  - image_templates:
-      - netbirdio/signal:{{ .Version }}-arm
-      - ghcr.io/netbirdio/signal:{{ .Version }}-arm
-    ids:
-      - netbird-signal
-    goarch: arm
-    goarm: 6
-    use: buildx
-    dockerfile: signal/Dockerfile
-    build_flag_templates:
-      - "--platform=linux/arm"
-      - "--label=org.opencontainers.image.created={{.Date}}"
-      - "--label=org.opencontainers.image.title={{.ProjectName}}"
-      - "--label=org.opencontainers.image.version={{.Version}}"
-      - "--label=org.opencontainers.image.revision={{.FullCommit}}"
-      - "--label=org.opencontainers.image.source=https://github.com/netbirdio/{{.ProjectName}}"
-      - "--label=maintainer=dev@netbird.io"
-  - image_templates:
-      - netbirdio/management:{{ .Version }}-amd64
-      - ghcr.io/netbirdio/management:{{ .Version }}-amd64
-    ids:
-      - netbird-mgmt
-    goarch: amd64
-    use: buildx
-    dockerfile: management/Dockerfile
-    build_flag_templates:
-      - "--platform=linux/amd64"
-      - "--label=org.opencontainers.image.created={{.Date}}"
-      - "--label=org.opencontainers.image.title={{.ProjectName}}"
-      - "--label=org.opencontainers.image.version={{.Version}}"
-      - "--label=org.opencontainers.image.revision={{.FullCommit}}"
-      - "--label=org.opencontainers.image.source=https://github.com/netbirdio/{{.ProjectName}}"
-      - "--label=maintainer=dev@netbird.io"
-  - image_templates:
-      - netbirdio/management:{{ .Version }}-arm64v8
-      - ghcr.io/netbirdio/management:{{ .Version }}-arm64v8
-    ids:
-      - netbird-mgmt
-    goarch: arm64
-    use: buildx
-    dockerfile: management/Dockerfile
-    build_flag_templates:
-      - "--platform=linux/arm64"
-      - "--label=org.opencontainers.image.created={{.Date}}"
-      - "--label=org.opencontainers.image.title={{.ProjectName}}"
-      - "--label=org.opencontainers.image.version={{.Version}}"
-      - "--label=org.opencontainers.image.revision={{.FullCommit}}"
-      - "--label=org.opencontainers.image.source=https://github.com/netbirdio/{{.ProjectName}}"
-      - "--label=maintainer=dev@netbird.io"
-  - image_templates:
-      - netbirdio/management:{{ .Version }}-arm
-      - ghcr.io/netbirdio/management:{{ .Version }}-arm
-    ids:
-      - netbird-mgmt
-    goarch: arm
-    goarm: 6
-    use: buildx
-    dockerfile: management/Dockerfile
-    build_flag_templates:
-      - "--platform=linux/arm"
-      - "--label=org.opencontainers.image.created={{.Date}}"
-      - "--label=org.opencontainers.image.title={{.ProjectName}}"
-      - "--label=org.opencontainers.image.version={{.Version}}"
-      - "--label=org.opencontainers.image.revision={{.FullCommit}}"
-      - "--label=org.opencontainers.image.source=https://github.com/netbirdio/{{.ProjectName}}"
-      - "--label=maintainer=dev@netbird.io"
-  - image_templates:
-      - netbirdio/management:{{ .Version }}-debug-amd64
-      - ghcr.io/netbirdio/management:{{ .Version }}-debug-amd64
-    ids:
-      - netbird-mgmt
-    goarch: amd64
-    use: buildx
-    dockerfile: management/Dockerfile.debug
-    build_flag_templates:
-      - "--platform=linux/amd64"
-      - "--label=org.opencontainers.image.created={{.Date}}"
-      - "--label=org.opencontainers.image.title={{.ProjectName}}"
-      - "--label=org.opencontainers.image.version={{.Version}}"
-      - "--label=org.opencontainers.image.revision={{.FullCommit}}"
-      - "--label=org.opencontainers.image.source=https://github.com/netbirdio/{{.ProjectName}}"
-      - "--label=maintainer=dev@netbird.io"
-  - image_templates:
-      - netbirdio/management:{{ .Version }}-debug-arm64v8
-      - ghcr.io/netbirdio/management:{{ .Version }}-debug-arm64v8
-    ids:
-      - netbird-mgmt
-    goarch: arm64
-    use: buildx
-    dockerfile: management/Dockerfile.debug
-    build_flag_templates:
-      - "--platform=linux/arm64"
-      - "--label=org.opencontainers.image.created={{.Date}}"
-      - "--label=org.opencontainers.image.title={{.ProjectName}}"
-      - "--label=org.opencontainers.image.version={{.Version}}"
-      - "--label=org.opencontainers.image.revision={{.FullCommit}}"
-      - "--label=org.opencontainers.image.source=https://github.com/netbirdio/{{.ProjectName}}"
-      - "--label=maintainer=dev@netbird.io"
-
-  - image_templates:
-      - netbirdio/management:{{ .Version }}-debug-arm
-      - ghcr.io/netbirdio/management:{{ .Version }}-debug-arm
-    ids:
-      - netbird-mgmt
-    goarch: arm
-    goarm: 6
-    use: buildx
-    dockerfile: management/Dockerfile.debug
-    build_flag_templates:
-      - "--platform=linux/arm"
-      - "--label=org.opencontainers.image.created={{.Date}}"
-      - "--label=org.opencontainers.image.title={{.ProjectName}}"
-      - "--label=org.opencontainers.image.version={{.Version}}"
-      - "--label=org.opencontainers.image.revision={{.FullCommit}}"
-      - "--label=org.opencontainers.image.source=https://github.com/netbirdio/{{.ProjectName}}"
-      - "--label=maintainer=dev@netbird.io"
-  - image_templates:
-      - netbirdio/upload:{{ .Version }}-amd64
-      - ghcr.io/netbirdio/upload:{{ .Version }}-amd64
-    ids:
-      - netbird-upload
-    goarch: amd64
-    use: buildx
-    dockerfile: upload-server/Dockerfile
-    build_flag_templates:
-      - "--platform=linux/amd64"
-      - "--label=org.opencontainers.image.created={{.Date}}"
-      - "--label=org.opencontainers.image.title={{.ProjectName}}"
-      - "--label=org.opencontainers.image.version={{.Version}}"
-      - "--label=org.opencontainers.image.revision={{.FullCommit}}"
-      - "--label=org.opencontainers.image.source=https://github.com/netbirdio/{{.ProjectName}}"
-      - "--label=maintainer=dev@netbird.io"
-  - image_templates:
-      - netbirdio/upload:{{ .Version }}-arm64v8
-      - ghcr.io/netbirdio/upload:{{ .Version }}-arm64v8
-    ids:
-      - netbird-upload
-    goarch: arm64
-    use: buildx
-    dockerfile: upload-server/Dockerfile
-    build_flag_templates:
-      - "--platform=linux/arm64"
-      - "--label=org.opencontainers.image.created={{.Date}}"
-      - "--label=org.opencontainers.image.title={{.ProjectName}}"
-      - "--label=org.opencontainers.image.version={{.Version}}"
-      - "--label=org.opencontainers.image.revision={{.FullCommit}}"
-      - "--label=org.opencontainers.image.source=https://github.com/netbirdio/{{.ProjectName}}"
-      - "--label=maintainer=dev@netbird.io"
-  - image_templates:
-      - netbirdio/upload:{{ .Version }}-arm
-      - ghcr.io/netbirdio/upload:{{ .Version }}-arm
-    ids:
-      - netbird-upload
-    goarch: arm
-    goarm: 6
-    use: buildx
-    dockerfile: upload-server/Dockerfile
-    build_flag_templates:
-      - "--platform=linux/arm"
-      - "--label=org.opencontainers.image.created={{.Date}}"
-      - "--label=org.opencontainers.image.title={{.ProjectName}}"
-      - "--label=org.opencontainers.image.version={{.Version}}"
-      - "--label=org.opencontainers.image.revision={{.FullCommit}}"
-      - "--label=org.opencontainers.image.source=https://github.com/netbirdio/{{.ProjectName}}"
-      - "--label=maintainer=dev@netbird.io"
-  - image_templates:
-      - netbirdio/netbird-server:{{ .Version }}-amd64
-      - ghcr.io/netbirdio/netbird-server:{{ .Version }}-amd64
-    ids:
-      - netbird-server
-    goarch: amd64
-    use: buildx
-    dockerfile: combined/Dockerfile
-    build_flag_templates:
-      - "--platform=linux/amd64"
-      - "--label=org.opencontainers.image.created={{.Date}}"
-      - "--label=org.opencontainers.image.title={{.ProjectName}}"
-      - "--label=org.opencontainers.image.version={{.Version}}"
-      - "--label=org.opencontainers.image.revision={{.FullCommit}}"
-      - "--label=org.opencontainers.image.source=https://github.com/netbirdio/{{.ProjectName}}"
-      - "--label=maintainer=dev@netbird.io"
-  - image_templates:
-      - netbirdio/netbird-server:{{ .Version }}-arm64v8
-      - ghcr.io/netbirdio/netbird-server:{{ .Version }}-arm64v8
-    ids:
-      - netbird-server
-    goarch: arm64
-    use: buildx
-    dockerfile: combined/Dockerfile
-    build_flag_templates:
-      - "--platform=linux/arm64"
-      - "--label=org.opencontainers.image.created={{.Date}}"
-      - "--label=org.opencontainers.image.title={{.ProjectName}}"
-      - "--label=org.opencontainers.image.version={{.Version}}"
-      - "--label=org.opencontainers.image.revision={{.FullCommit}}"
-      - "--label=org.opencontainers.image.source=https://github.com/netbirdio/{{.ProjectName}}"
-      - "--label=maintainer=dev@netbird.io"
-  - image_templates:
-      - netbirdio/netbird-server:{{ .Version }}-arm
-      - ghcr.io/netbirdio/netbird-server:{{ .Version }}-arm
-    ids:
-      - netbird-server
-    goarch: arm
-    goarm: 6
-    use: buildx
-    dockerfile: combined/Dockerfile
-    build_flag_templates:
-      - "--platform=linux/arm"
-      - "--label=org.opencontainers.image.created={{.Date}}"
-      - "--label=org.opencontainers.image.title={{.ProjectName}}"
-      - "--label=org.opencontainers.image.version={{.Version}}"
-      - "--label=org.opencontainers.image.revision={{.FullCommit}}"
-      - "--label=org.opencontainers.image.source=https://github.com/netbirdio/{{.ProjectName}}"
-      - "--label=maintainer=dev@netbird.io"
-  - image_templates:
-      - netbirdio/reverse-proxy:{{ .Version }}-amd64
-      - ghcr.io/netbirdio/reverse-proxy:{{ .Version }}-amd64
-    ids:
-      - netbird-proxy
-    goarch: amd64
-    use: buildx
-    dockerfile: proxy/Dockerfile
-    build_flag_templates:
-      - "--platform=linux/amd64"
-      - "--label=org.opencontainers.image.created={{.Date}}"
-      - "--label=org.opencontainers.image.title={{.ProjectName}}"
-      - "--label=org.opencontainers.image.version={{.Version}}"
-      - "--label=org.opencontainers.image.revision={{.FullCommit}}"
-      - "--label=org.opencontainers.image.source=https://github.com/netbirdio/{{.ProjectName}}"
-      - "--label=maintainer=dev@netbird.io"
-  - image_templates:
-      - netbirdio/reverse-proxy:{{ .Version }}-arm64v8
-      - ghcr.io/netbirdio/reverse-proxy:{{ .Version }}-arm64v8
-    ids:
-      - netbird-proxy
-    goarch: arm64
-    use: buildx
-    dockerfile: proxy/Dockerfile
-    build_flag_templates:
-      - "--platform=linux/arm64"
-      - "--label=org.opencontainers.image.created={{.Date}}"
-      - "--label=org.opencontainers.image.title={{.ProjectName}}"
-      - "--label=org.opencontainers.image.version={{.Version}}"
-      - "--label=org.opencontainers.image.revision={{.FullCommit}}"
-      - "--label=org.opencontainers.image.source=https://github.com/netbirdio/{{.ProjectName}}"
-      - "--label=maintainer=dev@netbird.io"
-  - image_templates:
-      - netbirdio/reverse-proxy:{{ .Version }}-arm
-      - ghcr.io/netbirdio/reverse-proxy:{{ .Version }}-arm
-    ids:
-      - netbird-proxy
-    goarch: arm
-    goarm: 6
-    use: buildx
-    dockerfile: proxy/Dockerfile
-    build_flag_templates:
-      - "--platform=linux/arm"
-      - "--label=org.opencontainers.image.created={{.Date}}"
-      - "--label=org.opencontainers.image.title={{.ProjectName}}"
-      - "--label=org.opencontainers.image.version={{.Version}}"
-      - "--label=org.opencontainers.image.revision={{.FullCommit}}"
-      - "--label=org.opencontainers.image.source=https://github.com/netbirdio/{{.ProjectName}}"
-      - "--label=maintainer=dev@netbird.io"
-docker_manifests:
-  - name_template: netbirdio/netbird:{{ .Version }}
-    image_templates:
-      - netbirdio/netbird:{{ .Version }}-arm64v8
-      - netbirdio/netbird:{{ .Version }}-arm
-      - netbirdio/netbird:{{ .Version }}-amd64
-
-  - name_template: netbirdio/netbird:latest
-    image_templates:
-      - netbirdio/netbird:{{ .Version }}-arm64v8
-      - netbirdio/netbird:{{ .Version }}-arm
-      - netbirdio/netbird:{{ .Version }}-amd64
-
-  - name_template: netbirdio/netbird:{{ .Version }}-rootless
-    image_templates:
-      - netbirdio/netbird:{{ .Version }}-rootless-arm64v8
-      - netbirdio/netbird:{{ .Version }}-rootless-arm
-      - netbirdio/netbird:{{ .Version }}-rootless-amd64
-
-  - name_template: netbirdio/netbird:rootless-latest
-    image_templates:
-      - netbirdio/netbird:{{ .Version }}-rootless-arm64v8
-      - netbirdio/netbird:{{ .Version }}-rootless-arm
-      - netbirdio/netbird:{{ .Version }}-rootless-amd64
-
-  - name_template: netbirdio/relay:{{ .Version }}
-    image_templates:
-      - netbirdio/relay:{{ .Version }}-arm64v8
-      - netbirdio/relay:{{ .Version }}-arm
-      - netbirdio/relay:{{ .Version }}-amd64
-
-  - name_template: netbirdio/relay:latest
-    image_templates:
-      - netbirdio/relay:{{ .Version }}-arm64v8
-      - netbirdio/relay:{{ .Version }}-arm
-      - netbirdio/relay:{{ .Version }}-amd64
-
-  - name_template: netbirdio/signal:{{ .Version }}
-    image_templates:
-      - netbirdio/signal:{{ .Version }}-arm64v8
-      - netbirdio/signal:{{ .Version }}-arm
-      - netbirdio/signal:{{ .Version }}-amd64
-
-  - name_template: netbirdio/signal:latest
-    image_templates:
-      - netbirdio/signal:{{ .Version }}-arm64v8
-      - netbirdio/signal:{{ .Version }}-arm
-      - netbirdio/signal:{{ .Version }}-amd64
-
-  - name_template: netbirdio/management:{{ .Version }}
-    image_templates:
-      - netbirdio/management:{{ .Version }}-arm64v8
-      - netbirdio/management:{{ .Version }}-arm
-      - netbirdio/management:{{ .Version }}-amd64
-
-  - name_template: netbirdio/management:latest
-    image_templates:
-      - netbirdio/management:{{ .Version }}-arm64v8
-      - netbirdio/management:{{ .Version }}-arm
-      - netbirdio/management:{{ .Version }}-amd64
-
-  - name_template: netbirdio/management:debug-latest
-    image_templates:
-      - netbirdio/management:{{ .Version }}-debug-arm64v8
-      - netbirdio/management:{{ .Version }}-debug-arm
-      - netbirdio/management:{{ .Version }}-debug-amd64
-  - name_template: netbirdio/upload:{{ .Version }}
-    image_templates:
-      - netbirdio/upload:{{ .Version }}-arm64v8
-      - netbirdio/upload:{{ .Version }}-arm
-      - netbirdio/upload:{{ .Version }}-amd64
-
-  - name_template: netbirdio/upload:latest
-    image_templates:
-      - netbirdio/upload:{{ .Version }}-arm64v8
-      - netbirdio/upload:{{ .Version }}-arm
-      - netbirdio/upload:{{ .Version }}-amd64
-
-  - name_template: netbirdio/netbird-server:{{ .Version }}
-    image_templates:
-      - netbirdio/netbird-server:{{ .Version }}-arm64v8
-      - netbirdio/netbird-server:{{ .Version }}-arm
-      - netbirdio/netbird-server:{{ .Version }}-amd64
-
-  - name_template: netbirdio/netbird-server:latest
-    image_templates:
-      - netbirdio/netbird-server:{{ .Version }}-arm64v8
-      - netbirdio/netbird-server:{{ .Version }}-arm
-      - netbirdio/netbird-server:{{ .Version }}-amd64
-
-  - name_template: ghcr.io/netbirdio/netbird:{{ .Version }}
-    image_templates:
-      - ghcr.io/netbirdio/netbird:{{ .Version }}-arm64v8
-      - ghcr.io/netbirdio/netbird:{{ .Version }}-arm
-      - ghcr.io/netbirdio/netbird:{{ .Version }}-amd64
-
-  - name_template: ghcr.io/netbirdio/netbird:latest
-    image_templates:
-      - ghcr.io/netbirdio/netbird:{{ .Version }}-arm64v8
-      - ghcr.io/netbirdio/netbird:{{ .Version }}-arm
-      - ghcr.io/netbirdio/netbird:{{ .Version }}-amd64
-
-  - name_template: ghcr.io/netbirdio/netbird:{{ .Version }}-rootless
-    image_templates:
-      - ghcr.io/netbirdio/netbird:{{ .Version }}-rootless-arm64v8
-      - ghcr.io/netbirdio/netbird:{{ .Version }}-rootless-arm
-      - ghcr.io/netbirdio/netbird:{{ .Version }}-rootless-amd64
-
-  - name_template: ghcr.io/netbirdio/netbird:rootless-latest
-    image_templates:
-      - ghcr.io/netbirdio/netbird:{{ .Version }}-rootless-arm64v8
-      - ghcr.io/netbirdio/netbird:{{ .Version }}-rootless-arm
-      - ghcr.io/netbirdio/netbird:{{ .Version }}-rootless-amd64
-
-  - name_template: ghcr.io/netbirdio/relay:{{ .Version }}
-    image_templates:
-      - ghcr.io/netbirdio/relay:{{ .Version }}-arm64v8
-      - ghcr.io/netbirdio/relay:{{ .Version }}-arm
-      - ghcr.io/netbirdio/relay:{{ .Version }}-amd64
-
-  - name_template: ghcr.io/netbirdio/relay:latest
-    image_templates:
-      - ghcr.io/netbirdio/relay:{{ .Version }}-arm64v8
-      - ghcr.io/netbirdio/relay:{{ .Version }}-arm
-      - ghcr.io/netbirdio/relay:{{ .Version }}-amd64
-
-  - name_template: ghcr.io/netbirdio/signal:{{ .Version }}
-    image_templates:
-      - ghcr.io/netbirdio/signal:{{ .Version }}-arm64v8
-      - ghcr.io/netbirdio/signal:{{ .Version }}-arm
-      - ghcr.io/netbirdio/signal:{{ .Version }}-amd64
-
-  - name_template: ghcr.io/netbirdio/signal:latest
-    image_templates:
-      - ghcr.io/netbirdio/signal:{{ .Version }}-arm64v8
-      - ghcr.io/netbirdio/signal:{{ .Version }}-arm
-      - ghcr.io/netbirdio/signal:{{ .Version }}-amd64
-
-  - name_template: ghcr.io/netbirdio/management:{{ .Version }}
-    image_templates:
-      - ghcr.io/netbirdio/management:{{ .Version }}-arm64v8
-      - ghcr.io/netbirdio/management:{{ .Version }}-arm
-      - ghcr.io/netbirdio/management:{{ .Version }}-amd64
-
-  - name_template: ghcr.io/netbirdio/management:latest
-    image_templates:
-      - ghcr.io/netbirdio/management:{{ .Version }}-arm64v8
-      - ghcr.io/netbirdio/management:{{ .Version }}-arm
-      - ghcr.io/netbirdio/management:{{ .Version }}-amd64
-
-  - name_template: ghcr.io/netbirdio/management:debug-latest
-    image_templates:
-      - ghcr.io/netbirdio/management:{{ .Version }}-debug-arm64v8
-      - ghcr.io/netbirdio/management:{{ .Version }}-debug-arm
-      - ghcr.io/netbirdio/management:{{ .Version }}-debug-amd64
-
-  - name_template: ghcr.io/netbirdio/upload:{{ .Version }}
-    image_templates:
-      - ghcr.io/netbirdio/upload:{{ .Version }}-arm64v8
-      - ghcr.io/netbirdio/upload:{{ .Version }}-arm
-      - ghcr.io/netbirdio/upload:{{ .Version }}-amd64
-
-  - name_template: ghcr.io/netbirdio/upload:latest
-    image_templates:
-      - ghcr.io/netbirdio/upload:{{ .Version }}-arm64v8
-      - ghcr.io/netbirdio/upload:{{ .Version }}-arm
-      - ghcr.io/netbirdio/upload:{{ .Version }}-amd64
-
-  - name_template: ghcr.io/netbirdio/netbird-server:{{ .Version }}
-    image_templates:
-      - ghcr.io/netbirdio/netbird-server:{{ .Version }}-arm64v8
-      - ghcr.io/netbirdio/netbird-server:{{ .Version }}-arm
-      - ghcr.io/netbirdio/netbird-server:{{ .Version }}-amd64
-
-  - name_template: ghcr.io/netbirdio/netbird-server:latest
-    image_templates:
-      - ghcr.io/netbirdio/netbird-server:{{ .Version }}-arm64v8
-      - ghcr.io/netbirdio/netbird-server:{{ .Version }}-arm
-      - ghcr.io/netbirdio/netbird-server:{{ .Version }}-amd64
-
-  - name_template: netbirdio/reverse-proxy:{{ .Version }}
-    image_templates:
-      - netbirdio/reverse-proxy:{{ .Version }}-arm64v8
-      - netbirdio/reverse-proxy:{{ .Version }}-arm
-      - netbirdio/reverse-proxy:{{ .Version }}-amd64
-
-  - name_template: netbirdio/reverse-proxy:latest
-    image_templates:
-      - netbirdio/reverse-proxy:{{ .Version }}-arm64v8
-      - netbirdio/reverse-proxy:{{ .Version }}-arm
-      - netbirdio/reverse-proxy:{{ .Version }}-amd64
-
-  - name_template: ghcr.io/netbirdio/reverse-proxy:{{ .Version }}
-    image_templates:
-      - ghcr.io/netbirdio/reverse-proxy:{{ .Version }}-arm64v8
-      - ghcr.io/netbirdio/reverse-proxy:{{ .Version }}-arm
-      - ghcr.io/netbirdio/reverse-proxy:{{ .Version }}-amd64
-
-  - name_template: ghcr.io/netbirdio/reverse-proxy:latest
-    image_templates:
-      - ghcr.io/netbirdio/reverse-proxy:{{ .Version }}-arm64v8
-      - ghcr.io/netbirdio/reverse-proxy:{{ .Version }}-arm
-      - ghcr.io/netbirdio/reverse-proxy:{{ .Version }}-amd64
 
 brews:
   - ids:
       - default
+    skip_upload: "{{ .Env.SKIP_PUBLISH }}"
     repository:
       owner: netbirdio
       name: homebrew-tap
@@ -902,6 +440,7 @@ brews:
 
 uploads:
   - name: debian
+    skip: "{{ .Env.SKIP_PUBLISH }}"
     ids:
       - netbird_deb
     mode: archive
@@ -910,6 +449,7 @@ uploads:
     method: PUT
 
   - name: yum
+    skip: "{{ .Env.SKIP_PUBLISH }}"
     ids:
       - netbird_rpm
     mode: archive
@@ -922,9 +462,13 @@ checksum:
     - glob: ./infrastructure_files/getting-started-with-zitadel.sh
     - glob: ./release_files/install.sh
     - glob: ./infrastructure_files/getting-started.sh
+    - glob: ./infrastructure_files/getting-started-enterprise.sh
+    - glob: ./infrastructure_files/migrate-to-enterprise.sh
 
 release:
   extra_files:
     - glob: ./infrastructure_files/getting-started-with-zitadel.sh
     - glob: ./release_files/install.sh
     - glob: ./infrastructure_files/getting-started.sh
+    - glob: ./infrastructure_files/getting-started-enterprise.sh
+    - glob: ./infrastructure_files/migrate-to-enterprise.sh

.goreleaser_ui.yaml

@@ -1,5 +1,6 @@
 version: 2
-
+env:
+  - SKIP_PUBLISH={{ if index .Env "SKIP_PUBLISH" }}{{ .Env.SKIP_PUBLISH }}{{ else }}true{{ end }}
 project_name: netbird-ui
 builds:
   - id: netbird-ui
@@ -101,6 +102,7 @@ nfpms:
 
 uploads:
   - name: debian
+    skip: "{{ .Env.SKIP_PUBLISH }}"
     ids:
       - netbird_ui_deb
     mode: archive
@@ -109,6 +111,7 @@ uploads:
     method: PUT
 
   - name: yum
+    skip: "{{ .Env.SKIP_PUBLISH }}"
     ids:
       - netbird_ui_rpm
     mode: archive

Makefile

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

README.md

@@ -37,6 +37,11 @@
   </strong>
 </p>
 
+> ### 🤖 NetBird Agent Network (Beta)
+> Identity-aware access control for AI agents — keyless access to LLM APIs and private
+> resources over the encrypted NetBird tunnel. See [`agent-network/`](agent-network/) or
+> read the docs at **[netbird.ai](https://netbird.ai)**.
+
 **NetBird combines a configuration-free peer-to-peer private network and a centralized access control system in a single platform, making it easy to create secure private networks for your organization or home.**
 
 **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.

agent-network/README.md

@@ -0,0 +1,39 @@
+# NetBird Agent Network
+
+Agent Network is NetBird's access control layer for AI agents and the people who run
+them. It gives every agent a real identity, tied to your identity provider (IdP), and
+governs what it can reach — the LLM APIs and AI gateways it can call, and the internal
+resources it can access. Traffic flows only over the encrypted NetBird tunnel, scoped by
+policy, with no API keys to leak.
+
+> **Beta.** Agent Network is open source and can be self-hosted on your own
+> infrastructure.
+
+## How it works
+
+Agent Network is built on two existing NetBird capabilities:
+
+- **Overlay network** — the encrypted WireGuard mesh between peers.
+- **Reverse proxy** — a NetBird peer that terminates LLM requests, establishes the
+  caller's identity, evaluates policies/limits/guardrails, injects the upstream provider
+  key server-side, forwards to the API or gateway, and records usage.
+
+LLM traffic is routed through the proxy's identity-aware pipeline, while internal
+resources (databases, internal APIs, self-hosted models) are reached directly over
+peer-to-peer WireGuard tunnels, governed by the same identities and access policies.
+
+## Where the code lives
+
+There is no separate "agent-network" service — it reuses the reverse-proxy and management
+components:
+
+- [`proxy/`](../proxy) — the NetBird reverse proxy that serves the agent network endpoint
+  and runs the per-request middleware pipeline.
+- [`management/internals/modules/reverseproxy/`](../management/internals/modules/reverseproxy)
+  — the management-side control plane: providers, policies, guardrails, limits, routing,
+  and usage/access logs.
+
+## Documentation
+
+Full documentation, architecture, and quickstart:
+**https://docs.netbird.io/agent-network**

client/Dockerfile

@@ -4,7 +4,7 @@
 #   sudo podman build -t localhost/netbird:latest -f client/Dockerfile --ignorefile .dockerignore-client .
 #   sudo podman run --rm -it --cap-add={BPF,NET_ADMIN,NET_RAW} localhost/netbird:latest
 
-FROM alpine:3.23.3
+FROM alpine:3.24
 # iproute2: busybox doesn't display ip rules properly
 RUN apk add --no-cache \
     bash \
@@ -21,7 +21,7 @@ ENV \
     NB_ENTRYPOINT_SERVICE_TIMEOUT="30"
 
 ENTRYPOINT [ "/usr/local/bin/netbird-entrypoint.sh" ]
-
+ARG TARGETPLATFORM
-ARG NETBIRD_BINARY=netbird
+ARG NETBIRD_BINARY=$TARGETPLATFORM/netbird
 COPY client/netbird-entrypoint.sh /usr/local/bin/netbird-entrypoint.sh
 COPY "${NETBIRD_BINARY}"  /usr/local/bin/netbird

client/Dockerfile-rootless

@@ -4,7 +4,7 @@
 #   podman build -t localhost/netbird:latest -f client/Dockerfile --ignorefile .dockerignore-client .
 #   podman run --rm -it --cap-add={BPF,NET_ADMIN,NET_RAW} localhost/netbird:latest
 
-FROM alpine:3.22.0
+FROM alpine:3.24
 
 RUN apk add --no-cache \
       bash \
@@ -27,7 +27,7 @@ ENV \
     NB_ENTRYPOINT_SERVICE_TIMEOUT="30"
 
 ENTRYPOINT [ "/usr/local/bin/netbird-entrypoint.sh" ]
-
+ARG TARGETPLATFORM
-ARG NETBIRD_BINARY=netbird
+ARG NETBIRD_BINARY=$TARGETPLATFORM/netbird
 COPY client/netbird-entrypoint.sh /usr/local/bin/netbird-entrypoint.sh
 COPY "${NETBIRD_BINARY}"  /usr/local/bin/netbird

client/android/profile_manager.go

@@ -6,7 +6,6 @@ import (
 	"fmt"
 	"os"
 	"path/filepath"
-	"strings"
 
 	log "github.com/sirupsen/logrus"
 
@@ -24,6 +23,7 @@ const (
 
 // Profile represents a profile for gomobile
 type Profile struct {
+	ID       string
 	Name     string
 	IsActive bool
 }
@@ -53,10 +53,10 @@ func (p *ProfileArray) Get(i int) *Profile {
 ├── state.json                                  ← Default profile state
 ├── active_profile.json                         ← Active profile tracker (JSON with Name + Username)
 └── profiles/                                   ← Subdirectory for non-default profiles
-    ├── work.json                              ← Work profile config
+    ├── work.json                              			← Legacy work profile config
-    ├── work.state.json                        ← Work profile state
+    ├── work.state.json                        			← Legacy work profile state
-    ├── personal.json                          ← Personal profile config
+    ├── 4c5f5c8198c3989cffb5b5394f5a7ae0.json  			← ID profile config
-    └── personal.state.json                    ← Personal profile state
+    ├── 4c5f5c8198c3989cffb5b5394f5a7ae0.state.json ← ID profile state
 */
 
 // ProfileManager manages profiles for Android
@@ -99,6 +99,7 @@ func (pm *ProfileManager) ListProfiles() (*ProfileArray, error) {
 	var profiles []*Profile
 	for _, p := range internalProfiles {
 		profiles = append(profiles, &Profile{
+			ID:       p.ID.String(),
 			Name:     p.Name,
 			IsActive: p.IsActive,
 		})
@@ -108,55 +109,65 @@ func (pm *ProfileManager) ListProfiles() (*ProfileArray, error) {
 }
 
 // GetActiveProfile returns the currently active profile name
-func (pm *ProfileManager) GetActiveProfile() (string, error) {
+func (pm *ProfileManager) GetActiveProfile() (*Profile, error) {
 	// Use ServiceManager to stay consistent with ListProfiles
 	// ServiceManager uses active_profile.json
 	activeState, err := pm.serviceMgr.GetActiveProfileState()
 	if err != nil {
-		return "", fmt.Errorf("failed to get active profile: %w", err)
+		return nil, fmt.Errorf("failed to get active profile: %w", err)
 	}
-	return activeState.Name, nil
+
+	// ActiveProfileState only stores the ID (and username), not the display
+	// name. Resolve the ID to the full profile so callers get the real Name.
+	prof, err := pm.serviceMgr.ResolveProfile(activeState.ID.String(), androidUsername)
+	if err != nil {
+		return nil, fmt.Errorf("failed to resolve active profile %q: %w", activeState.ID, err)
+	}
+	return &Profile{ID: prof.ID.String(), Name: prof.Name, IsActive: true}, nil
 }
 
 // SwitchProfile switches to a different profile
-func (pm *ProfileManager) SwitchProfile(profileName string) error {
+func (pm *ProfileManager) SwitchProfile(id string) error {
 	// Use ServiceManager to stay consistent with ListProfiles
 	// ServiceManager uses active_profile.json
 	err := pm.serviceMgr.SetActiveProfileState(&profilemanager.ActiveProfileState{
-		Name:     profileName,
+		ID:       profilemanager.ID(id),
 		Username: androidUsername,
 	})
 	if err != nil {
 		return fmt.Errorf("failed to switch profile: %w", err)
 	}
 
-	log.Infof("switched to profile: %s", profileName)
+	log.Infof("switched to profile: %s", id)
 	return nil
 }
 
 // AddProfile creates a new profile
 func (pm *ProfileManager) AddProfile(profileName string) error {
 	// Use ServiceManager (creates profile in profiles/ directory)
-	if err := pm.serviceMgr.AddProfile(profileName, androidUsername); err != nil {
+	profile, err := pm.serviceMgr.AddProfile(profileName, androidUsername)
+	if err != nil {
 		return fmt.Errorf("failed to add profile: %w", err)
 	}
 
-	log.Infof("created new profile: %s", profileName)
+	log.Infof("created new profile: %s", profile.ID)
 	return nil
 }
 
 // LogoutProfile logs out from a profile (clears authentication)
-func (pm *ProfileManager) LogoutProfile(profileName string) error {
+func (pm *ProfileManager) LogoutProfile(id string) error {
-	profileName = sanitizeProfileName(profileName)
+	configPath, err := pm.getProfileConfigPath(id)
-
-	configPath, err := pm.getProfileConfigPath(profileName)
 	if err != nil {
 		return err
 	}
 
+	if !profilemanager.IsValidProfileFilenameStem(profilemanager.ID(id)) {
+		return fmt.Errorf("id '%s' is not valid", id)
+	}
+
 	// Check if profile exists
 	if _, err := os.Stat(configPath); os.IsNotExist(err) {
-		return fmt.Errorf("profile '%s' does not exist", profileName)
+		return fmt.Errorf("profile '%s' does not exist", id)
 	}
 
 	// Read current config using internal profilemanager
@@ -174,53 +185,57 @@ func (pm *ProfileManager) LogoutProfile(profileName string) error {
 		return fmt.Errorf("failed to save config: %w", err)
 	}
 
-	log.Infof("logged out from profile: %s", profileName)
+	log.Infof("logged out from profile: %s", id)
 	return nil
 }
 
 // RemoveProfile deletes a profile
-func (pm *ProfileManager) RemoveProfile(profileName string) error {
+func (pm *ProfileManager) RemoveProfile(id string) error {
 	// Use ServiceManager (removes profile from profiles/ directory)
-	if err := pm.serviceMgr.RemoveProfile(profileName, androidUsername); err != nil {
+	if err := pm.serviceMgr.RemoveProfile(profilemanager.ID(id), androidUsername); err != nil {
 		return fmt.Errorf("failed to remove profile: %w", err)
 	}
 
-	log.Infof("removed profile: %s", profileName)
+	log.Infof("removed profile: %s", id)
 	return nil
 }
 
 // getProfileConfigPath returns the config file path for a profile
 // This is needed for Android-specific path handling (netbird.cfg for default profile)
-func (pm *ProfileManager) getProfileConfigPath(profileName string) (string, error) {
+func (pm *ProfileManager) getProfileConfigPath(id string) (string, error) {
-	if profileName == "" || profileName == profilemanager.DefaultProfileName {
+	if !profilemanager.IsValidProfileFilenameStem(profilemanager.ID(id)) {
+		return "", fmt.Errorf("id %q is not valid", id)
+	}
+
+	if id == profilemanager.DefaultProfileName {
 		// Android uses netbird.cfg for default profile instead of default.json
 		// Default profile is stored in root configDir, not in profiles/
 		return filepath.Join(pm.configDir, defaultConfigFilename), nil
 	}
 
-	// Non-default profiles are stored in profiles subdirectory
-	// This matches the Java Preferences.java expectation
-	profileName = sanitizeProfileName(profileName)
 	profilesDir := filepath.Join(pm.configDir, profilesSubdir)
-	return filepath.Join(profilesDir, profileName+".json"), nil
+	return filepath.Join(profilesDir, id+".json"), nil
 }
 
-// GetConfigPath returns the config file path for a given profile
+// GetConfigPath returns the config file path for a given profile id
 // Java should call this instead of constructing paths with Preferences.configFile()
-func (pm *ProfileManager) GetConfigPath(profileName string) (string, error) {
+func (pm *ProfileManager) GetConfigPath(id string) (string, error) {
-	return pm.getProfileConfigPath(profileName)
+	return pm.getProfileConfigPath(id)
 }
 
 // GetStateFilePath returns the state file path for a given profile
 // Java should call this instead of constructing paths with Preferences.stateFile()
-func (pm *ProfileManager) GetStateFilePath(profileName string) (string, error) {
+func (pm *ProfileManager) GetStateFilePath(id string) (string, error) {
-	if profileName == "" || profileName == profilemanager.DefaultProfileName {
+	if id == "" || id == profilemanager.DefaultProfileName {
 		return filepath.Join(pm.configDir, "state.json"), nil
 	}
 
-	profileName = sanitizeProfileName(profileName)
+	if !profilemanager.IsValidProfileFilenameStem(profilemanager.ID(id)) {
+		return "", fmt.Errorf("id %q is not valid", id)
+	}
+
 	profilesDir := filepath.Join(pm.configDir, profilesSubdir)
-	return filepath.Join(profilesDir, profileName+".state.json"), nil
+	return filepath.Join(profilesDir, id+".state.json"), nil
 }
 
 // GetActiveConfigPath returns the config file path for the currently active profile
@@ -230,7 +245,7 @@ func (pm *ProfileManager) GetActiveConfigPath() (string, error) {
 	if err != nil {
 		return "", fmt.Errorf("failed to get active profile: %w", err)
 	}
-	return pm.GetConfigPath(activeProfile)
+	return pm.GetConfigPath(activeProfile.ID)
 }
 
 // GetActiveStateFilePath returns the state file path for the currently active profile
@@ -240,18 +255,5 @@ func (pm *ProfileManager) GetActiveStateFilePath() (string, error) {
 	if err != nil {
 		return "", fmt.Errorf("failed to get active profile: %w", err)
 	}
-	return pm.GetStateFilePath(activeProfile)
+	return pm.GetStateFilePath(activeProfile.ID)
-}
-
-// sanitizeProfileName removes invalid characters from profile name
-func sanitizeProfileName(name string) string {
-	// Keep only alphanumeric, underscore, and hyphen
-	var result strings.Builder
-	for _, r := range name {
-		if (r >= 'a' && r <= 'z') || (r >= 'A' && r <= 'Z') ||
-			(r >= '0' && r <= '9') || r == '_' || r == '-' {
-			result.WriteRune(r)
-		}
-	}
-	return result.String()
 }

client/cmd/debug.go

@@ -3,12 +3,14 @@ package cmd
 import (
 	"context"
 	"fmt"
+	"os/user"
 	"strings"
 	"time"
 
 	log "github.com/sirupsen/logrus"
 	"github.com/spf13/cobra"
 	"google.golang.org/grpc/status"
+	"google.golang.org/protobuf/encoding/protojson"
 	"google.golang.org/protobuf/types/known/durationpb"
 
 	"github.com/netbirdio/netbird/client/internal"
@@ -85,6 +87,73 @@ var persistenceCmd = &cobra.Command{
 	RunE:    setSyncResponsePersistence,
 }
 
+var debugConfigCmd = &cobra.Command{
+	Use:     "config",
+	Example: "  netbird debug config",
+	Short:   "Dump the effective configuration",
+	Long:    "Prints the daemon's resolved configuration (after applying defaults, file, env, CLI input, and MDM policy overrides) as JSON. Includes the list of MDM-managed fields.",
+	RunE:    debugConfigDump,
+}
+
+// debugConfigDump implements `netbird debug config`. It resolves the
+// active profile, queries the daemon for the effective configuration
+// via GetConfig, and prints the resulting GetConfigResponse as JSON
+// (via protojson with EmitUnpopulated=true so the output is stable
+// across runs and includes zero-valued fields).
+//
+// Useful for verifying MDM enforcement end-to-end: the response's
+// mDMManagedFields array is the single source of truth for "which
+// fields is the daemon currently enforcing from the MDM source", and
+// every config field side-by-side with that list confirms the merge
+// result. Secrets in the response (e.g. PreSharedKey) are already
+// redacted by the daemon-side handler.
+func debugConfigDump(cmd *cobra.Command, _ []string) error {
+	pm := profilemanager.NewProfileManager()
+	activeProf, err := pm.GetActiveProfile()
+	if err != nil {
+		return fmt.Errorf("get active profile: %v", err)
+	}
+	currUser, err := user.Current()
+	if err != nil {
+		return fmt.Errorf("get current user: %v", err)
+	}
+
+	conn, err := getClient(cmd)
+	if err != nil {
+		return err
+	}
+	defer func() {
+		if err := conn.Close(); err != nil {
+			log.Errorf(errCloseConnection, err)
+		}
+	}()
+
+	client := proto.NewDaemonServiceClient(conn)
+	resp, err := client.GetConfig(cmd.Context(), &proto.GetConfigRequest{
+		ProfileName: string(activeProf.ID),
+		Username:    currUser.Username,
+	})
+	if err != nil {
+		return fmt.Errorf("failed to get config: %v", status.Convert(err).Message())
+	}
+
+	// Use protojson so well-known fields render correctly; emit defaults so
+	// the operator sees every field even when zero/empty.
+	m := protojson.MarshalOptions{Multiline: true, Indent: "  ", EmitUnpopulated: true}
+	out, err := m.Marshal(resp)
+	if err != nil {
+		return fmt.Errorf("marshal config: %w", err)
+	}
+	cmd.Println(string(out))
+	return nil
+}
+
+// debugBundle requests the daemon to create a debug bundle and prints
+// the resulting local file path and, if uploaded, the uploaded file
+// key. It uses the package flags (anonymize, system info, log file
+// count, CLI version, optional upload URL) to configure the bundle
+// request. Returns an error if the RPC fails or if the daemon reports
+// an upload failure reason.
 func debugBundle(cmd *cobra.Command, _ []string) error {
 	conn, err := getClient(cmd)
 	if err != nil {

client/cmd/kubernetes.go

@@ -0,0 +1,301 @@
+package cmd
+
+import (
+	"context"
+	"crypto/tls"
+	"encoding/json"
+	"errors"
+	"fmt"
+	"io"
+	"net"
+	"net/http"
+	"net/url"
+	"os"
+	"path/filepath"
+	"slices"
+	"strings"
+
+	"github.com/goccy/go-yaml"
+	log "github.com/sirupsen/logrus"
+	"github.com/spf13/cobra"
+
+	"github.com/netbirdio/netbird/client/proto"
+)
+
+const (
+	KubernetesDNSSuffix = "netbird-kubeapi-proxy"
+)
+
+var kubernetesCmd = &cobra.Command{
+	Use:   "kubernetes",
+	Short: "Kubernetes cluster commands.",
+	Long:  "Kubernetes cluster commands.",
+}
+
+var kubernetesListCmd = &cobra.Command{
+	Use:   "list",
+	RunE:  kubernetesList,
+	Short: "List Kubernetes clusters.",
+	Long:  "List Kubernetes clusters by discovering NetBird peers running netbird-kubeapi-proxy.",
+}
+
+var kubernetesWriteKubeconfigCmd = &cobra.Command{
+	Use:   "write-kubeconfig",
+	RunE:  kubernetesWriteKubeconfig,
+	Args:  cobra.ExactArgs(1),
+	Short: "Write kubeconfig for a Kubernetes cluster.",
+	Long:  "Updates kubeconfig in place to allow token-less access to the Kubernetes cluster through NetBird.",
+}
+
+func init() {
+	kubernetesWriteKubeconfigCmd.Flags().String("kubeconfig", "", "path to kubeconfig file")
+}
+
+func kubernetesList(cmd *cobra.Command, _ []string) error {
+	conn, err := getClient(cmd)
+	if err != nil {
+		return err
+	}
+	defer conn.Close()
+	client := proto.NewDaemonServiceClient(conn)
+	statusResp, err := client.Status(cmd.Context(), &proto.StatusRequest{GetFullPeerStatus: true})
+	if err != nil {
+		return err
+	}
+
+	kcs, err := getKubernetesClusters(cmd.Context(), statusResp.FullStatus.Peers, "")
+	if err != nil {
+		return err
+	}
+	if len(kcs) == 0 {
+		cmd.Println("No Kubernetes clusters available.")
+		return nil
+	}
+	cmd.Println("Available Kubernetes clusters:")
+	for _, k := range kcs {
+		cmd.Printf("\n  - Name: %s\n    FQDN: %s\n    Version: %s\n", k.name, k.url.Host, k.version)
+	}
+	return nil
+}
+
+func kubernetesWriteKubeconfig(cmd *cobra.Command, args []string) error {
+	kubeconfigPath, err := resolveKubeconfigPath(cmd)
+	if err != nil {
+		return err
+	}
+
+	conn, err := getClient(cmd)
+	if err != nil {
+		return err
+	}
+	defer conn.Close()
+	client := proto.NewDaemonServiceClient(conn)
+	statusResp, err := client.Status(cmd.Context(), &proto.StatusRequest{GetFullPeerStatus: true})
+	if err != nil {
+		return err
+	}
+
+	clusterName := args[0]
+	kcs, err := getKubernetesClusters(cmd.Context(), statusResp.FullStatus.Peers, clusterName)
+	if err != nil {
+		return err
+	}
+	if len(kcs) == 0 {
+		return fmt.Errorf("kubernetes cluster named %s not found", clusterName)
+	}
+	if len(kcs) > 1 {
+		return fmt.Errorf("too many Kubernetes clusters returned")
+	}
+	err = writeKubeconfig(kubeconfigPath, kcs[0])
+	if err != nil {
+		return err
+	}
+	return nil
+}
+
+type kubernetesCluster struct {
+	name    string
+	url     *url.URL
+	version string
+}
+
+func getKubernetesClusters(ctx context.Context, peers []*proto.PeerState, nameFilter string) ([]kubernetesCluster, error) {
+	transport := http.DefaultTransport.(*http.Transport).Clone()
+	transport.TLSClientConfig = &tls.Config{
+		InsecureSkipVerify: true,
+	}
+	httpClient := &http.Client{
+		Transport: transport,
+	}
+	resolver := net.Resolver{
+		// Required so both DNS records are returned.
+		// https://github.com/golang/go/issues/17093
+		PreferGo: true,
+	}
+
+	kcs := []kubernetesCluster{}
+	attempted := map[string]struct{}{}
+	for _, peer := range peers {
+		fqdns, err := resolver.LookupAddr(ctx, peer.IP)
+		if err != nil {
+			return nil, err
+		}
+		for _, fqdn := range fqdns {
+			if _, ok := attempted[fqdn]; ok {
+				continue
+			}
+			attempted[fqdn] = struct{}{}
+			comps := strings.Split(fqdn, ".")
+			if len(comps) < 2 {
+				continue
+			}
+			if comps[1] != KubernetesDNSSuffix {
+				continue
+			}
+			if nameFilter != "" && nameFilter != comps[0] {
+				continue
+			}
+			clusterURL, clusterVersion, err := fingerprintClusters(ctx, httpClient, fqdn)
+			if err != nil {
+				log.Debugf("could not fingerprint Kubernetes cluster %s %q", fqdn, err)
+				continue
+			}
+			kc := kubernetesCluster{
+				name:    comps[0],
+				url:     clusterURL,
+				version: clusterVersion,
+			}
+			if nameFilter != "" {
+				return []kubernetesCluster{kc}, nil
+			}
+			kcs = append(kcs, kc)
+		}
+	}
+	return kcs, nil
+}
+
+func fingerprintClusters(ctx context.Context, httpClient *http.Client, fqdn string) (*url.URL, string, error) {
+	clusterURL, err := url.Parse("https://" + fqdn)
+	if err != nil {
+		return nil, "", err
+	}
+	versionURL, err := clusterURL.Parse("/version")
+	if err != nil {
+		return nil, "", err
+	}
+	req, err := http.NewRequestWithContext(ctx, http.MethodGet, versionURL.String(), nil)
+	if err != nil {
+		return nil, "", err
+	}
+	resp, err := httpClient.Do(req)
+	if err != nil {
+		return nil, "", err
+	}
+	defer resp.Body.Close()
+	if resp.StatusCode != http.StatusOK {
+		return nil, "", fmt.Errorf("expected %d response but got %s", http.StatusOK, resp.Status)
+	}
+	b, err := io.ReadAll(resp.Body)
+	if err != nil {
+		return nil, "", err
+	}
+	versionData := map[string]string{}
+	err = json.Unmarshal(b, &versionData)
+	if err != nil {
+		return nil, "", err
+	}
+	version, ok := versionData["gitVersion"]
+	if !ok {
+		return nil, "", errors.New("no version found in response")
+	}
+	return clusterURL, version, nil
+}
+
+func resolveKubeconfigPath(cmd *cobra.Command) (string, error) {
+	if cmd.Flags().Changed("kubeconfig") {
+		path, err := cmd.Flags().GetString("kubeconfig")
+		if err != nil {
+			return "", err
+		}
+		return path, nil
+	}
+	if env := os.Getenv("KUBECONFIG"); env != "" {
+		return env, nil
+	}
+	home, err := os.UserHomeDir()
+	if err != nil {
+		return "", fmt.Errorf("could not determine home directory: %w", err)
+	}
+	return filepath.Join(home, ".kube", "config"), nil
+}
+
+func writeKubeconfig(kubeconfigPath string, kc kubernetesCluster) error {
+	b, err := os.ReadFile(kubeconfigPath)
+	if err != nil && !errors.Is(err, os.ErrNotExist) {
+		return err
+	}
+	var cfg map[string]any
+	if err := yaml.Unmarshal(b, &cfg); err != nil {
+		return err
+	}
+	if cfg == nil {
+		cfg = map[string]any{
+			"apiVersion": "v1",
+			"kind":       "Config",
+		}
+	}
+
+	cfg["clusters"] = appendWithName(cfg["clusters"], map[string]any{
+		"name": kc.name,
+		"cluster": map[string]any{
+			"server":                   kc.url.String(),
+			"insecure-skip-tls-verify": true,
+		},
+	})
+	cfg["users"] = appendWithName(cfg["users"], map[string]any{
+		"name": "netbird",
+		"user": map[string]any{
+			"token": "none",
+		},
+	})
+	cfg["contexts"] = appendWithName(cfg["contexts"], map[string]any{
+		"name": kc.name,
+		"context": map[string]any{
+			"cluster":   kc.name,
+			"user":      "netbird",
+			"namespace": "default",
+		},
+	})
+	cfg["current-context"] = kc.name
+
+	out, err := yaml.Marshal(cfg)
+	if err != nil {
+		return err
+	}
+	if err := os.WriteFile(kubeconfigPath, out, 0o600); err != nil {
+		return err
+	}
+	return nil
+}
+
+func appendWithName(data any, add map[string]any) any {
+	if data == nil {
+		return []any{add}
+	}
+	v, ok := data.([]any)
+	if !ok {
+		return []any{add}
+	}
+	i := slices.IndexFunc(v, func(item any) bool {
+		m, ok := item.(map[string]any)
+		if !ok {
+			return false
+		}
+		return m["name"] == add["name"]
+	})
+	if i == -1 {
+		return append(v, add)
+	}
+	v[i] = add
+	return v
+}

client/cmd/kubernetes_test.go

@@ -0,0 +1,120 @@
+package cmd
+
+import (
+	"net/http"
+	"net/http/httptest"
+	"net/url"
+	"os"
+	"path/filepath"
+	"testing"
+
+	"github.com/spf13/cobra"
+	"github.com/stretchr/testify/require"
+)
+
+func TestFingerprintClusters(t *testing.T) {
+	t.Parallel()
+
+	srv := httptest.NewTLSServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
+		//nolint: errcheck
+		w.Write([]byte(`{"gitVersion": "foobar"}`))
+	}))
+	defer srv.Close()
+
+	clusterURL, clusterVersion, err := fingerprintClusters(t.Context(), srv.Client(), srv.Listener.Addr().String())
+	require.NoError(t, err)
+	require.Equal(t, srv.URL, clusterURL.String())
+	require.Equal(t, "foobar", clusterVersion)
+}
+
+func TestResolveKubeconfigPath(t *testing.T) {
+	home, err := os.UserHomeDir()
+	if err != nil {
+		t.Fatalf("could not determine home directory: %v", err)
+	}
+	defaultPath := filepath.Join(home, ".kube", "config")
+	path, err := resolveKubeconfigPath(&cobra.Command{})
+	require.NoError(t, err)
+	require.Equal(t, defaultPath, path)
+
+	flagPath := "flag-path"
+	cmd := &cobra.Command{}
+	cmd.Flags().String("kubeconfig", "", "")
+	err = cmd.Flags().Set("kubeconfig", flagPath)
+	require.NoError(t, err)
+	path, err = resolveKubeconfigPath(cmd)
+	require.NoError(t, err)
+	require.Equal(t, flagPath, path)
+
+	envPath := "env-path"
+	t.Setenv("KUBECONFIG", envPath)
+	path, err = resolveKubeconfigPath(&cobra.Command{})
+	require.NoError(t, err)
+	require.Equal(t, envPath, path)
+}
+
+func TestWriteKubeconfig(t *testing.T) {
+	t.Parallel()
+
+	tests := []struct {
+		name     string
+		existing string
+	}{
+		{
+			name: "empty file",
+		},
+		{
+			name: "existing content",
+			existing: `apiVersion: v1
+clusters:
+- cluster:
+    insecure-skip-tls-verify: true
+    server: https://foobar.com
+  name: foo
+current-context: test
+kind: Config
+users: []
+`,
+		},
+	}
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			t.Parallel()
+
+			kubeconfigPath := filepath.Join(t.TempDir(), "config")
+			err := os.WriteFile(kubeconfigPath, []byte(tt.existing), 0o644)
+			require.NoError(t, err)
+
+			kc := kubernetesCluster{
+				name: "foo",
+				url:  &url.URL{Scheme: "https", Host: "example.com"},
+			}
+			err = writeKubeconfig(kubeconfigPath, kc)
+			require.NoError(t, err)
+
+			b, err := os.ReadFile(kubeconfigPath)
+			require.NoError(t, err)
+			expected := `apiVersion: v1
+clusters:
+- cluster:
+    insecure-skip-tls-verify: true
+    server: https://example.com
+  name: foo
+contexts:
+- context:
+    cluster: foo
+    namespace: default
+    user: netbird
+  name: foo
+current-context: foo
+kind: Config
+users:
+- name: netbird
+  user:
+    token: none
+`
+			require.Equal(t, expected, string(b))
+		})
+	}
+
+}

client/cmd/login.go

@@ -96,17 +96,19 @@ func doDaemonLogin(ctx context.Context, cmd *cobra.Command, providedSetupKey str
 		dnsLabelsReq = dnsLabelsValidated.ToSafeStringList()
 	}
 
+	handle := activeProf.ID.String()
+
 	loginRequest := proto.LoginRequest{
 		SetupKey:            providedSetupKey,
 		ManagementUrl:       managementURL,
 		IsUnixDesktopClient: isUnixRunningDesktop(),
 		Hostname:            hostName,
 		DnsLabels:           dnsLabelsReq,
-		ProfileName:         &activeProf.Name,
+		ProfileName:         &handle,
 		Username:            &username,
 	}
 
-	profileState, err := pm.GetProfileState(activeProf.Name)
+	profileState, err := pm.GetProfileState(activeProf.ID)
 	if err != nil {
 		log.Debugf("failed to get profile state for login hint: %v", err)
 	} else if profileState.Email != "" {
@@ -170,14 +172,13 @@ func getActiveProfile(ctx context.Context, pm *profilemanager.ProfileManager, pr
 	return activeProf, nil
 }
 
-func switchProfileOnDaemon(ctx context.Context, pm *profilemanager.ProfileManager, profileName string, username string) error {
+func switchProfileOnDaemon(ctx context.Context, pm *profilemanager.ProfileManager, handle string, username string) error {
-	err := switchProfile(context.Background(), profileName, username)
+	resolvedID, err := switchProfile(ctx, handle, username)
 	if err != nil {
 		return fmt.Errorf("switch profile on daemon: %v", err)
 	}
 
-	err = pm.SwitchProfile(profileName)
+	if err := pm.SwitchProfile(resolvedID); err != nil {
-	if err != nil {
 		return fmt.Errorf("switch profile: %v", err)
 	}
 
@@ -205,11 +206,15 @@ func switchProfileOnDaemon(ctx context.Context, pm *profilemanager.ProfileManage
 	return nil
 }
 
-func switchProfile(ctx context.Context, profileName string, username string) error {
+// switchProfile asks the daemon to switch to the profile identified by
+// handle (a name, ID, or unique ID prefix). Returns the resolved profile
+// ID so the caller can update the local active-profile state without
+// re-resolving the handle.
+func switchProfile(ctx context.Context, handle string, username string) (profilemanager.ID, error) {
 	conn, err := DialClientGRPCServer(ctx, daemonAddr)
 	if err != nil {
 		//nolint
-		return fmt.Errorf("failed to connect to daemon error: %v\n"+
+		return "", 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)
 	}
@@ -217,15 +222,15 @@ func switchProfile(ctx context.Context, profileName string, username string) err
 
 	client := proto.NewDaemonServiceClient(conn)
 
-	_, err = client.SwitchProfile(ctx, &proto.SwitchProfileRequest{
+	resp, err := client.SwitchProfile(ctx, &proto.SwitchProfileRequest{
-		ProfileName: &profileName,
+		ProfileName: &handle,
 		Username:    &username,
 	})
 	if err != nil {
-		return fmt.Errorf("switch profile failed: %v", err)
+		return "", fmt.Errorf("switch profile failed: %w", err)
 	}
 
-	return nil
+	return profilemanager.ID(resp.Id), nil
 }
 
 func doForegroundLogin(ctx context.Context, cmd *cobra.Command, setupKey string, activeProf *profilemanager.Profile) error {
@@ -249,7 +254,7 @@ func doForegroundLogin(ctx context.Context, cmd *cobra.Command, setupKey string,
 		return fmt.Errorf("read config file %s: %v", configFilePath, err)
 	}
 
-	err = foregroundLogin(ctx, cmd, config, setupKey, activeProf.Name)
+	err = foregroundLogin(ctx, cmd, config, setupKey, activeProf.ID)
 	if err != nil {
 		return fmt.Errorf("foreground login failed: %v", err)
 	}
@@ -277,7 +282,7 @@ func handleSSOLogin(ctx context.Context, cmd *cobra.Command, loginResp *proto.Lo
 	return nil
 }
 
-func foregroundLogin(ctx context.Context, cmd *cobra.Command, config *profilemanager.Config, setupKey, profileName string) error {
+func foregroundLogin(ctx context.Context, cmd *cobra.Command, config *profilemanager.Config, setupKey string, profileID profilemanager.ID) error {
 	authClient, err := auth.NewAuth(ctx, config.PrivateKey, config.ManagementURL, config)
 	if err != nil {
 		return fmt.Errorf("failed to create auth client: %v", err)
@@ -291,7 +296,7 @@ func foregroundLogin(ctx context.Context, cmd *cobra.Command, config *profileman
 
 	jwtToken := ""
 	if setupKey == "" && needsLogin {
-		tokenInfo, err := foregroundGetTokenInfo(ctx, cmd, config, profileName)
+		tokenInfo, err := foregroundGetTokenInfo(ctx, cmd, config, profileID)
 		if err != nil {
 			return fmt.Errorf("interactive sso login failed: %v", err)
 		}
@@ -306,10 +311,10 @@ func foregroundLogin(ctx context.Context, cmd *cobra.Command, config *profileman
 	return nil
 }
 
-func foregroundGetTokenInfo(ctx context.Context, cmd *cobra.Command, config *profilemanager.Config, profileName string) (*auth.TokenInfo, error) {
+func foregroundGetTokenInfo(ctx context.Context, cmd *cobra.Command, config *profilemanager.Config, profileID profilemanager.ID) (*auth.TokenInfo, error) {
 	hint := ""
 	pm := profilemanager.NewProfileManager()
-	profileState, err := pm.GetProfileState(profileName)
+	profileState, err := pm.GetProfileState(profileID)
 	if err != nil {
 		log.Debugf("failed to get profile state for login hint: %v", err)
 	} else if profileState.Email != "" {

client/cmd/login_test.go

@@ -27,7 +27,7 @@ func TestLogin(t *testing.T) {
 	profilemanager.ActiveProfileStatePath = tempDir + "/active_profile.json"
 	sm := profilemanager.ServiceManager{}
 	err = sm.SetActiveProfileState(&profilemanager.ActiveProfileState{
-		Name:     "default",
+		ID:       "default",
 		Username: currUser.Username,
 	})
 	if err != nil {

client/cmd/profile.go

@@ -2,11 +2,16 @@ package cmd
 
 import (
 	"context"
+	"errors"
 	"fmt"
 	"os/user"
+	"strings"
+	"text/tabwriter"
 	"time"
 
 	"github.com/spf13/cobra"
+	"google.golang.org/grpc/codes"
+	gstatus "google.golang.org/grpc/status"
 
 	"github.com/netbirdio/netbird/client/internal"
 	"github.com/netbirdio/netbird/client/internal/profilemanager"
@@ -14,6 +19,8 @@ import (
 	"github.com/netbirdio/netbird/util"
 )
 
+var profileListShowID bool
+
 var profileCmd = &cobra.Command{
 	Use:   "profile",
 	Short: "Manage NetBird client profiles",
@@ -31,27 +38,40 @@ var profileListCmd = &cobra.Command{
 var profileAddCmd = &cobra.Command{
 	Use:   "add <profile_name>",
 	Short: "Add a new profile",
-	Long:  `Add a new profile to the NetBird client. The profile name must be unique.`,
+	Long:  `Add a new profile. Profile name is free-form, a unique ID is generated for the on-disk config file.`,
 	Args:  cobra.ExactArgs(1),
 	RunE:  addProfileFunc,
 }
 
+var profileRenameCmd = &cobra.Command{
+	Use:   "rename <profile> <new_profile_name>",
+	Short: "Renames an existing profile",
+	Long:  `Renames an existing profile (by a name, ID, or unique ID prefix). Profile name is free-form.`,
+	Args:  cobra.ExactArgs(2),
+	RunE:  renameProfileFunc,
+}
+
 var profileRemoveCmd = &cobra.Command{
-	Use:   "remove <profile_name>",
+	Use:     "remove <profile>",
-	Short: "Remove a profile",
+	Short:   "Remove a profile",
-	Long:  `Remove a profile from the NetBird client. The profile must not be inactive.`,
+	Long:    `Remove a profile by name, ID, or unique ID prefix.`,
-	Args:  cobra.ExactArgs(1),
+	Aliases: []string{"rm"},
-	RunE:  removeProfileFunc,
+	Args:    cobra.ExactArgs(1),
+	RunE:    removeProfileFunc,
 }
 
 var profileSelectCmd = &cobra.Command{
-	Use:   "select <profile_name>",
+	Use:   "select <profile>",
 	Short: "Select a profile",
-	Long:  `Make the specified profile active. This will switch the client to use the selected profile's configuration.`,
+	Long:  `Make the specified profile active. Accepts a name, ID, or unique ID prefix.`,
 	Args:  cobra.ExactArgs(1),
 	RunE:  selectProfileFunc,
 }
 
+func init() {
+	profileListCmd.Flags().BoolVar(&profileListShowID, "show-id", false, "show the profile ID column")
+}
+
 func setupCmd(cmd *cobra.Command) error {
 	SetFlagsFromEnvVars(rootCmd)
 	SetFlagsFromEnvVars(cmd)
@@ -65,6 +85,7 @@ func setupCmd(cmd *cobra.Command) error {
 
 	return nil
 }
+
 func listProfilesFunc(cmd *cobra.Command, _ []string) error {
 	if err := setupCmd(cmd); err != nil {
 		return err
@@ -83,25 +104,33 @@ func listProfilesFunc(cmd *cobra.Command, _ []string) error {
 
 	daemonClient := proto.NewDaemonServiceClient(conn)
 
-	profiles, err := daemonClient.ListProfiles(cmd.Context(), &proto.ListProfilesRequest{
+	resp, err := daemonClient.ListProfiles(cmd.Context(), &proto.ListProfilesRequest{
 		Username: currUser.Username,
 	})
 	if err != nil {
 		return err
 	}
 
-	// list profiles, add a tick if the profile is active
+	tw := tabwriter.NewWriter(cmd.OutOrStdout(), 0, 0, 2, ' ', 0)
-	cmd.Println("Found", len(profiles.Profiles), "profiles:")
+	if profileListShowID {
-	for _, profile := range profiles.Profiles {
+		fmt.Fprintln(tw, "ID\tNAME\tACTIVE")
-		// use a cross to indicate the passive profiles
+	} else {
-		activeMarker := "✗"
+		fmt.Fprintln(tw, "NAME\tACTIVE")
-		if profile.IsActive {
-			activeMarker = "✓"
-		}
-		cmd.Println(activeMarker, profile.Name)
 	}
-
+	for _, profile := range resp.Profiles {
-	return nil
+		marker := ""
+		if profile.IsActive {
+			marker = "✓"
+		}
+		name := profilemanager.StripCtrlChars(profile.Name)
+		id := profilemanager.ID(profile.Id)
+		if profileListShowID {
+			fmt.Fprintf(tw, "%s\t%s\t%s\n", id.ShortID(), name, marker)
+		} else {
+			fmt.Fprintf(tw, "%s\t%s\n", name, marker)
+		}
+	}
+	return tw.Flush()
 }
 
 func addProfileFunc(cmd *cobra.Command, args []string) error {
@@ -109,6 +138,41 @@ func addProfileFunc(cmd *cobra.Command, args []string) error {
 		return err
 	}
 
+	currUser, err := user.Current()
+	if err != nil {
+		return fmt.Errorf("get current user: %w", err)
+	}
+
+	conn, err := DialClientGRPCServer(cmd.Context(), daemonAddr)
+	if err != nil {
+		return fmt.Errorf("connect to service CLI interface: %w", err)
+	}
+	defer conn.Close()
+
+	daemonClient := proto.NewDaemonServiceClient(conn)
+	profileName := args[0]
+
+	id, err := addProfileOnDaemon(cmd.Context(), daemonClient, profileName, currUser.Username)
+	if err != nil {
+		return err
+	}
+
+	dupCount, _ := countProfilesWithName(cmd.Context(), daemonClient, currUser.Username, profileName)
+	if dupCount > 1 {
+		cmd.Printf("Warning: %d other profile(s) already use the name %q.\n", dupCount-1, profileName)
+		cmd.Println("Use `netbird profile list --show-id` to disambiguate later.")
+	}
+
+	cmd.Printf("Profile added: %s  %s\n", id.ShortID(), profilemanager.StripCtrlChars(profileName))
+	return nil
+
+}
+
+func renameProfileFunc(cmd *cobra.Command, args []string) error {
+	if err := setupCmd(cmd); err != nil {
+		return err
+	}
+
 	conn, err := DialClientGRPCServer(cmd.Context(), daemonAddr)
 	if err != nil {
 		return fmt.Errorf("connect to service CLI interface: %w", err)
@@ -121,21 +185,43 @@ func addProfileFunc(cmd *cobra.Command, args []string) error {
 	}
 
 	daemonClient := proto.NewDaemonServiceClient(conn)
+	handle := args[0]
+	newProfilename := args[1]
 
-	profileName := args[0]
+	resp, err := daemonClient.RenameProfile(cmd.Context(), &proto.RenameProfileRequest{
-
+		Handle:         handle,
-	_, err = daemonClient.AddProfile(cmd.Context(), &proto.AddProfileRequest{
+		Username:       currUser.Username,
-		ProfileName: profileName,
+		NewProfileName: newProfilename,
-		Username:    currUser.Username,
 	})
 	if err != nil {
-		return err
+		return wrapAmbiguityError(err, handle)
 	}
 
-	cmd.Println("Profile added successfully:", profileName)
+	dupCount, _ := countProfilesWithName(cmd.Context(), daemonClient, currUser.Username, newProfilename)
+	if dupCount > 1 {
+		cmd.Printf("Warning: %d other profile(s) already use the name %q.\n", dupCount-1, newProfilename)
+		cmd.Println("Use `netbird profile list --show-id` to disambiguate later.")
+	}
+
+	cmd.Printf("Profile renamed from %s to %s\n", profilemanager.StripCtrlChars(resp.OldProfileName), profilemanager.StripCtrlChars(newProfilename))
+
 	return nil
 }
 
+func countProfilesWithName(ctx context.Context, c proto.DaemonServiceClient, username, name string) (int, error) {
+	resp, err := c.ListProfiles(ctx, &proto.ListProfilesRequest{Username: username})
+	if err != nil {
+		return 0, err
+	}
+	n := 0
+	for _, p := range resp.Profiles {
+		if p.Name == name {
+			n++
+		}
+	}
+	return n, nil
+}
+
 func removeProfileFunc(cmd *cobra.Command, args []string) error {
 	if err := setupCmd(cmd); err != nil {
 		return err
@@ -153,18 +239,17 @@ func removeProfileFunc(cmd *cobra.Command, args []string) error {
 	}
 
 	daemonClient := proto.NewDaemonServiceClient(conn)
+	handle := args[0]
 
-	profileName := args[0]
+	resp, err := daemonClient.RemoveProfile(cmd.Context(), &proto.RemoveProfileRequest{
-
+		ProfileName: handle,
-	_, err = daemonClient.RemoveProfile(cmd.Context(), &proto.RemoveProfileRequest{
-		ProfileName: profileName,
 		Username:    currUser.Username,
 	})
 	if err != nil {
-		return err
+		return wrapAmbiguityError(err, handle)
 	}
 
-	cmd.Println("Profile removed successfully:", profileName)
+	cmd.Printf("Profile removed: %s\n", resp.Id)
 	return nil
 }
 
@@ -174,7 +259,7 @@ func selectProfileFunc(cmd *cobra.Command, args []string) error {
 	}
 
 	profileManager := profilemanager.NewProfileManager()
-	profileName := args[0]
+	handle := args[0]
 
 	currUser, err := user.Current()
 	if err != nil {
@@ -191,32 +276,15 @@ func selectProfileFunc(cmd *cobra.Command, args []string) error {
 
 	daemonClient := proto.NewDaemonServiceClient(conn)
 
-	profiles, err := daemonClient.ListProfiles(ctx, &proto.ListProfilesRequest{
+	switchResp, err := daemonClient.SwitchProfile(ctx, &proto.SwitchProfileRequest{
-		Username: currUser.Username,
+		ProfileName: &handle,
+		Username:    &currUser.Username,
 	})
 	if err != nil {
-		return fmt.Errorf("list profiles: %w", err)
+		return wrapAmbiguityError(err, handle)
 	}
 
-	var profileExists bool
+	if err := profileManager.SwitchProfile(profilemanager.ID(switchResp.Id)); err != nil {
-
-	for _, profile := range profiles.Profiles {
-		if profile.Name == profileName {
-			profileExists = true
-			break
-		}
-	}
-
-	if !profileExists {
-		return fmt.Errorf("profile %s does not exist", profileName)
-	}
-
-	if err := switchProfile(cmd.Context(), profileName, currUser.Username); err != nil {
-		return err
-	}
-
-	err = profileManager.SwitchProfile(profileName)
-	if err != nil {
 		return err
 	}
 
@@ -231,6 +299,46 @@ func selectProfileFunc(cmd *cobra.Command, args []string) error {
 		}
 	}
 
-	cmd.Println("Profile switched successfully to:", profileName)
+	id := profilemanager.ID(switchResp.Id)
+	cmd.Printf("Profile switched to: %s\n", id.ShortID())
 	return nil
 }
+
+// wrapAmbiguityError turns the daemon's gRPC InvalidArgument errors
+// (which carry the resolver's message verbatim) into CLI-friendly text
+// that points the user at --show-id.
+func wrapAmbiguityError(err error, handle string) error {
+	if err == nil {
+		return nil
+	}
+	st, ok := gstatus.FromError(err)
+	if !ok {
+		return err
+	}
+	switch st.Code() {
+	case codes.InvalidArgument:
+		msg := st.Message()
+		if strings.Contains(msg, "ambiguous") {
+			return errors.New(msg + "\nRun `netbird profile list --show-id` to see IDs, then select by ID prefix:\n  netbird profile select|remove <id-prefix>")
+		}
+	case codes.NotFound:
+		return fmt.Errorf("profile %q not found", handle)
+	}
+	return err
+}
+
+// addProfileOnDaemon issues the AddProfile RPC on an existing daemon client
+// and returns the new profile's ID. It is the single entry point for profile
+// creation, shared by `netbird profile add` and the `netbird up --profile
+// <name>` auto-create path.
+func addProfileOnDaemon(ctx context.Context, client proto.DaemonServiceClient, profileName, username string) (profilemanager.ID, error) {
+	resp, err := client.AddProfile(ctx, &proto.AddProfileRequest{
+		ProfileName: profileName,
+		Username:    username,
+	})
+	if err != nil {
+		return "", fmt.Errorf("add profile failed: %w", err)
+	}
+
+	return profilemanager.ID(resp.Id), nil
+}

client/cmd/root.go

@@ -95,7 +95,9 @@ var (
 	}
 )
 
-// Execute executes the root command.
+// Execute runs the appropriate Cobra command for the CLI.
+// If the process is the update binary it delegates to updateCmd; otherwise it runs the root command.
+// It returns any error produced during command execution.
 func Execute() error {
 	if isUpdateBinary() {
 		return updateCmd.Execute()
@@ -103,6 +105,16 @@ func Execute() error {
 	return rootCmd.Execute()
 }
 
+// init initialises package-level defaults and configures the root
+// Cobra command tree. Sets platform-specific config / log directory
+// paths (including legacy Wiretrustee fallbacks) and a default daemon
+// address; registers persistent CLI flags (daemon address,
+// management / admin URLs, logging, setup key (file and inline,
+// mutually exclusive), preshared key, hostname, anonymise, config
+// path); attaches top-level and nested subcommands to the root
+// command; and registers `up`-specific persistent flags (external IP
+// maps, custom DNS resolver address, Rosenpass options, auto-connect
+// disabling, lazy connection).
 func init() {
 	defaultConfigPathDir = "/etc/netbird/"
 	defaultLogFileDir = "/var/log/netbird/"
@@ -168,10 +180,17 @@ func init() {
 	logCmd.AddCommand(logLevelCmd)
 	debugCmd.AddCommand(forCmd)
 	debugCmd.AddCommand(persistenceCmd)
+	debugCmd.AddCommand(debugConfigCmd)
+
+	// kubernetes commands
+	rootCmd.AddCommand(kubernetesCmd)
+	kubernetesCmd.AddCommand(kubernetesListCmd)
+	kubernetesCmd.AddCommand(kubernetesWriteKubeconfigCmd)
 
 	// profile commands
 	profileCmd.AddCommand(profileListCmd)
 	profileCmd.AddCommand(profileAddCmd)
+	profileCmd.AddCommand(profileRenameCmd)
 	profileCmd.AddCommand(profileRemoveCmd)
 	profileCmd.AddCommand(profileSelectCmd)
 

client/cmd/service_privileged_test.go

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

client/cmd/service_test.go

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

client/cmd/status.go

@@ -11,7 +11,6 @@ import (
 	"google.golang.org/grpc/status"
 
 	"github.com/netbirdio/netbird/client/internal"
-	"github.com/netbirdio/netbird/client/internal/profilemanager"
 	"github.com/netbirdio/netbird/client/proto"
 	nbstatus "github.com/netbirdio/netbird/client/status"
 	"github.com/netbirdio/netbird/util"
@@ -111,11 +110,10 @@ func statusFunc(cmd *cobra.Command, args []string) error {
 		return nil
 	}
 
-	pm := profilemanager.NewProfileManager()
+	// Resolve the active profile's display name via the daemon, which runs
-	var profName string
+	// as root and can read the per-user profile files. The local profile
-	if activeProf, err := pm.GetActiveProfile(); err == nil {
+	// manager only knows the active profile ID, not its display name.
-		profName = activeProf.Name
+	profName := getActiveProfileName(ctx)
-	}
 
 	var outputInformationHolder = nbstatus.ConvertToStatusOutputOverview(resp.GetFullStatus(), nbstatus.ConvertOptions{
 		Anonymize:            anonymizeFlag,
@@ -167,6 +165,25 @@ func getStatus(ctx context.Context, fullPeerStatus bool, shouldRunProbes bool) (
 	return resp, nil
 }
 
+// getActiveProfileName asks the daemon for the active profile's display
+// name. The daemon runs as root and can read the per-user profile files to
+// resolve the ID to its human-readable name. Returns an empty string on any
+// error so status output degrades gracefully.
+func getActiveProfileName(ctx context.Context) string {
+	conn, err := DialClientGRPCServer(ctx, daemonAddr)
+	if err != nil {
+		return ""
+	}
+	defer conn.Close()
+
+	resp, err := proto.NewDaemonServiceClient(conn).GetActiveProfile(ctx, &proto.GetActiveProfileRequest{})
+	if err != nil {
+		return ""
+	}
+
+	return resp.GetProfileName()
+}
+
 func parseFilters() error {
 	switch strings.ToLower(statusFilter) {
 	case "", "idle", "connecting", "connected":

client/cmd/up.go

@@ -128,16 +128,9 @@ func upFunc(cmd *cobra.Command, args []string) error {
 	var profileSwitched bool
 	// switch profile if provided
 	if profileName != "" {
-		err = switchProfile(cmd.Context(), profileName, username.Username)
+		if err := switchOrCreateProfile(cmd.Context(), pm, profileName, username.Username); err != nil {
-		if err != nil {
 			return fmt.Errorf("switch profile: %v", err)
 		}
-
-		err = pm.SwitchProfile(profileName)
-		if err != nil {
-			return fmt.Errorf("switch profile: %v", err)
-		}
-
 		profileSwitched = true
 	}
 
@@ -152,6 +145,52 @@ func upFunc(cmd *cobra.Command, args []string) error {
 	return runInDaemonMode(ctx, cmd, pm, activeProf, profileSwitched)
 }
 
+// switchOrCreateProfile switches the active profile to the one identified by
+// handle, creating it first when it does not exist yet. This restores the
+// pre-0.73 behaviour where `netbird up --profile <name>` auto-creates a
+// missing profile instead of failing.
+func switchOrCreateProfile(ctx context.Context, pm *profilemanager.ProfileManager, handle, username string) error {
+	resolvedID, err := switchProfile(ctx, handle, username)
+	if err != nil {
+		st, ok := gstatus.FromError(err)
+		if !ok || st.Code() != codes.NotFound {
+			return err
+		}
+		// Don't fail immediately on a create error: a concurrent run may
+		// have created the profile between the NotFound above and this
+		// call, in which case the retried switch still succeeds. Only
+		// surface the create error if the switch also fails.
+		_, createErr := createProfile(ctx, handle, username)
+		if resolvedID, err = switchProfile(ctx, handle, username); err != nil {
+			if createErr != nil {
+				return fmt.Errorf("create profile: %w", createErr)
+			}
+			return err
+		}
+	}
+
+	if err := pm.SwitchProfile(resolvedID); err != nil {
+		return err
+	}
+	return nil
+}
+
+// createProfile dials the daemon and creates a new profile with the given
+// display name, returning its generated ID. Use addProfileOnDaemon directly
+// when a daemon client is already available to reuse the connection.
+func createProfile(ctx context.Context, profileName, username string) (profilemanager.ID, error) {
+	conn, err := DialClientGRPCServer(ctx, daemonAddr)
+	if err != nil {
+		//nolint
+		return "", 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)
+	}
+	defer conn.Close()
+
+	return addProfileOnDaemon(ctx, proto.NewDaemonServiceClient(conn), profileName, username)
+}
+
 func runInForegroundMode(ctx context.Context, cmd *cobra.Command, activeProf *profilemanager.Profile) error {
 	// override the default profile filepath if provided
 	if configPath != "" {
@@ -190,7 +229,7 @@ func runInForegroundMode(ctx context.Context, cmd *cobra.Command, activeProf *pr
 
 	_, _ = profilemanager.UpdateOldManagementURL(ctx, config, configFilePath)
 
-	err = foregroundLogin(ctx, cmd, config, providedSetupKey, activeProf.Name)
+	err = foregroundLogin(ctx, cmd, config, providedSetupKey, activeProf.ID)
 	if err != nil {
 		return fmt.Errorf("foreground login failed: %v", err)
 	}
@@ -261,10 +300,10 @@ func runInDaemonMode(ctx context.Context, cmd *cobra.Command, pm *profilemanager
 	}
 
 	// set the new config
-	req := setupSetConfigReq(customDNSAddressConverted, cmd, activeProf.Name, username.Username)
+	req := setupSetConfigReq(customDNSAddressConverted, cmd, activeProf.ID.String(), username.Username)
 	if _, err := client.SetConfig(ctx, req); err != nil {
 		if st, ok := gstatus.FromError(err); ok && st.Code() == codes.Unavailable {
-			log.Warnf("setConfig method is not available in the daemon")
+			log.Warnf("setConfig method is not available in the daemon: %s", st.Message())
 		} else {
 			return fmt.Errorf("call service setConfig method: %v", err)
 		}
@@ -289,10 +328,11 @@ func doDaemonUp(ctx context.Context, cmd *cobra.Command, client proto.DaemonServ
 		return fmt.Errorf("setup login request: %v", err)
 	}
 
-	loginRequest.ProfileName = &activeProf.Name
+	profileID := activeProf.ID.String()
+	loginRequest.ProfileName = &profileID
 	loginRequest.Username = &username
 
-	profileState, err := pm.GetProfileState(activeProf.Name)
+	profileState, err := pm.GetProfileState(activeProf.ID)
 	if err != nil {
 		log.Debugf("failed to get profile state for login hint: %v", err)
 	} else if profileState.Email != "" {
@@ -329,7 +369,7 @@ func doDaemonUp(ctx context.Context, cmd *cobra.Command, client proto.DaemonServ
 	}
 
 	if _, err := client.Up(ctx, &proto.UpRequest{
-		ProfileName: &activeProf.Name,
+		ProfileName: &profileID,
 		Username:    &username,
 	}); err != nil {
 		return fmt.Errorf("call service up method: %v", err)

client/cmd/up_daemon_test.go

@@ -29,14 +29,14 @@ func TestUpDaemon(t *testing.T) {
 	}
 
 	sm := profilemanager.ServiceManager{}
-	err = sm.AddProfile("test1", currUser.Username)
+	created, err := sm.AddProfile("test1", currUser.Username)
 	if err != nil {
 		t.Fatalf("failed to add profile: %v", err)
 		return
 	}
 
 	err = sm.SetActiveProfileState(&profilemanager.ActiveProfileState{
-		Name:     "test1",
+		ID:       created.ID,
 		Username: currUser.Username,
 	})
 	if err != nil {

client/embed/embed.go

@@ -279,6 +279,12 @@ func (c *Client) Start(startCtx context.Context) error {
 
 	select {
 	case <-startCtx.Done():
+		// ConnectClient.Stop now cancels its own run context and waits for the
+		// run loop to tear the engine down, so this cancel() is no longer
+		// required to break the deadlock and could be removed. It is kept as a
+		// defensive belt-and-suspenders: cancelling the parent context first
+		// guarantees the run loop is unblocked even if Stop's contract regresses.
+		cancel()
 		if stopErr := client.Stop(); stopErr != nil {
 			return fmt.Errorf("stop error after context done. Stop error: %w. Context done: %w", stopErr, startCtx.Err())
 		}
@@ -442,8 +448,8 @@ func (c *Client) Expose(ctx context.Context, req ExposeRequest) (*ExposeSession,
 
 // IdentityForIP looks up a remote peer by its tunnel IP using the
 // embedded client's status recorder. Returns the peer's WireGuard public
-// key and FQDN. ok=false means the IP isn't in this client's peer
+// key and FQDN. ok=false means the IP doesn't belong to an active peer
-// roster — callers should treat that as "unknown peer".
+// — offline roster peers are treated as unknown, same as foreign IPs.
 func (c *Client) IdentityForIP(ip netip.Addr) (pubKey, fqdn string, ok bool) {
 	if !ip.IsValid() || c.recorder == nil {
 		return "", "", false

client/embed/embed_test.go

@@ -0,0 +1,168 @@
+package embed
+
+import (
+	"context"
+	"net"
+	"testing"
+	"time"
+
+	"github.com/golang/mock/gomock"
+	"github.com/stretchr/testify/require"
+	"google.golang.org/grpc"
+
+	"github.com/netbirdio/netbird/management/internals/controllers/network_map/controller"
+	"github.com/netbirdio/netbird/management/internals/controllers/network_map/update_channel"
+	"github.com/netbirdio/netbird/management/internals/modules/peers"
+	"github.com/netbirdio/netbird/management/internals/modules/peers/ephemeral/manager"
+	"github.com/netbirdio/netbird/management/internals/server/config"
+	nbgrpc "github.com/netbirdio/netbird/management/internals/shared/grpc"
+	mgmt "github.com/netbirdio/netbird/management/server"
+	"github.com/netbirdio/netbird/management/server/activity"
+	nbcache "github.com/netbirdio/netbird/management/server/cache"
+	"github.com/netbirdio/netbird/management/server/groups"
+	"github.com/netbirdio/netbird/management/server/integrations/integrated_validator/validator"
+	"github.com/netbirdio/netbird/management/server/integrations/port_forwarding"
+	"github.com/netbirdio/netbird/management/server/job"
+	"github.com/netbirdio/netbird/management/server/permissions"
+	"github.com/netbirdio/netbird/management/server/settings"
+	"github.com/netbirdio/netbird/management/server/store"
+	"github.com/netbirdio/netbird/management/server/telemetry"
+	"github.com/netbirdio/netbird/management/server/types"
+	mgmtProto "github.com/netbirdio/netbird/shared/management/proto"
+	"github.com/netbirdio/netbird/util"
+)
+
+const testSetupKey = "A2C8E62B-38F5-4553-B31E-DD66C696CEBB"
+
+// TestClientStartTimeoutRollback reproduces a deadlock between Engine.Start and
+// Engine.Stop. The signal endpoint accepts gRPC connections but never serves the
+// SignalExchange service, so Engine.Start parks in WaitStreamConnected while
+// holding the engine mutex. When the Start context expires, the rollback path
+// calls ConnectClient.Stop, which must not block forever acquiring that mutex.
+func TestClientStartTimeoutRollback(t *testing.T) {
+	signalAddr := startBlackholeSignal(t)
+	mgmAddr := startManagement(t, signalAddr)
+
+	wgPort := 0
+	client, err := New(Options{
+		DeviceName:    "embed-rollback-test",
+		SetupKey:      testSetupKey,
+		ManagementURL: "http://" + mgmAddr,
+		WireguardPort: &wgPort,
+	})
+	require.NoError(t, err, "embed client creation must succeed")
+
+	startCtx, cancel := context.WithTimeout(context.Background(), 5*time.Second)
+	defer cancel()
+
+	startErr := make(chan error, 1)
+	go func() {
+		startErr <- client.Start(startCtx)
+	}()
+
+	select {
+	case err := <-startErr:
+		require.ErrorIs(t, err, context.DeadlineExceeded)
+	case <-time.After(60 * time.Second):
+		t.Fatal("client.Start did not return after its context expired: Engine.Stop deadlocked against Engine.Start waiting for the signal stream")
+	}
+}
+
+// startBlackholeSignal starts a gRPC server without the SignalExchange service
+// registered. Connections succeed, but the signal stream can never be
+// established, which keeps Engine.Start parked in WaitStreamConnected.
+func startBlackholeSignal(t *testing.T) string {
+	t.Helper()
+
+	lis, err := net.Listen("tcp", "localhost:0")
+	require.NoError(t, err)
+
+	s := grpc.NewServer()
+	go func() {
+		if err := s.Serve(lis); err != nil {
+			t.Error(err)
+		}
+	}()
+	t.Cleanup(s.Stop)
+
+	return lis.Addr().String()
+}
+
+func startManagement(t *testing.T, signalAddr string) string {
+	t.Helper()
+
+	cfg := &config.Config{
+		Stuns:      []*config.Host{},
+		TURNConfig: &config.TURNConfig{},
+		Relay: &config.Relay{
+			Addresses:      []string{"127.0.0.1:1234"},
+			CredentialsTTL: util.Duration{Duration: time.Hour},
+			Secret:         "222222222222222222",
+		},
+		Signal: &config.Host{
+			Proto: "http",
+			URI:   signalAddr,
+		},
+		Datadir:    t.TempDir(),
+		HttpConfig: nil,
+	}
+
+	lis, err := net.Listen("tcp", "localhost:0")
+	require.NoError(t, err)
+
+	s := grpc.NewServer()
+
+	testStore, cleanUp, err := store.NewTestStoreFromSQL(context.Background(), "../testdata/store.sql", cfg.Datadir)
+	require.NoError(t, err)
+	t.Cleanup(cleanUp)
+
+	eventStore := &activity.InMemoryEventStore{}
+
+	permissionsManager := permissions.NewManager(testStore)
+	peersManager := peers.NewManager(testStore, permissionsManager)
+	jobManager := job.NewJobManager(nil, testStore, peersManager)
+
+	cacheStore, err := nbcache.NewStore(context.Background(), 100*time.Millisecond, 300*time.Millisecond, 100)
+	require.NoError(t, err)
+
+	iv, err := validator.NewIntegratedValidator(context.Background(), peersManager, nil, eventStore, cacheStore)
+	require.NoError(t, err)
+	metrics, err := telemetry.NewDefaultAppMetrics(context.Background())
+	require.NoError(t, err)
+
+	ctrl := gomock.NewController(t)
+	t.Cleanup(ctrl.Finish)
+	settingsMockManager := settings.NewMockManager(ctrl)
+	settingsMockManager.EXPECT().
+		GetSettings(gomock.Any(), gomock.Any(), gomock.Any()).
+		Return(&types.Settings{}, nil).
+		AnyTimes()
+	settingsMockManager.EXPECT().
+		GetExtraSettings(gomock.Any(), gomock.Any()).
+		Return(&types.ExtraSettings{}, nil).
+		AnyTimes()
+
+	groupsManager := groups.NewManagerMock()
+
+	updateManager := update_channel.NewPeersUpdateManager(metrics)
+	requestBuffer := mgmt.NewAccountRequestBuffer(context.Background(), testStore)
+	networkMapController := controller.NewController(context.Background(), testStore, metrics, updateManager, requestBuffer, mgmt.MockIntegratedValidator{}, settingsMockManager, "netbird.selfhosted", port_forwarding.NewControllerMock(), manager.NewEphemeralManager(testStore, peersManager), cfg)
+	accountManager, err := mgmt.BuildManager(context.Background(), cfg, testStore, networkMapController, jobManager, nil, "", eventStore, nil, false, iv, metrics, port_forwarding.NewControllerMock(), settingsMockManager, permissionsManager, false, cacheStore)
+	require.NoError(t, err)
+
+	secretsManager, err := nbgrpc.NewTimeBasedAuthSecretsManager(updateManager, cfg.TURNConfig, cfg.Relay, settingsMockManager, groupsManager)
+	require.NoError(t, err)
+
+	mgmtServer, err := nbgrpc.NewServer(cfg, accountManager, settingsMockManager, jobManager, secretsManager, nil, nil, &mgmt.MockIntegratedValidator{}, networkMapController, nil, nil)
+	require.NoError(t, err)
+	mgmtProto.RegisterManagementServiceServer(s, mgmtServer)
+
+	go func() {
+		if err := s.Serve(lis); err != nil {
+			t.Error(err)
+		}
+	}()
+	t.Cleanup(s.Stop)
+
+	return lis.Addr().String()
+}

client/firewall/iptables/acl_linux.go

@@ -3,6 +3,7 @@ package iptables
 import (
 	"errors"
 	"fmt"
+	"maps"
 	"net"
 	"slices"
 
@@ -421,12 +422,17 @@ func (m *aclManager) updateState() {
 	currentState.Lock()
 	defer currentState.Unlock()
 
+	// Clone the maps so the persisted state holds a private snapshot. The
+	// live maps keep being mutated by subsequent rule operations while the
+	// state manager marshals the state from its periodic-save goroutine.
+	// Sharing them by reference races the two and aborts the process with a
+	// concurrent map iteration and write.
 	if m.v6 {
-		currentState.ACLEntries6 = m.entries
+		currentState.ACLEntries6 = maps.Clone(m.entries)
-		currentState.ACLIPsetStore6 = m.ipsetStore
+		currentState.ACLIPsetStore6 = m.ipsetStore.clone()
 	} else {
-		currentState.ACLEntries = m.entries
+		currentState.ACLEntries = maps.Clone(m.entries)
-		currentState.ACLIPsetStore = m.ipsetStore
+		currentState.ACLIPsetStore = m.ipsetStore.clone()
 	}
 
 	if err := m.stateManager.UpdateState(currentState); err != nil {

client/firewall/iptables/manager_linux_test.go

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

client/firewall/iptables/router_linux.go

@@ -4,6 +4,7 @@ package iptables
 
 import (
 	"fmt"
+	"maps"
 	"net/netip"
 	"strconv"
 	"strings"
@@ -749,11 +750,17 @@ func (r *router) updateState() {
 	currentState.Lock()
 	defer currentState.Unlock()
 
+	// Clone the rule map so the persisted state holds a private snapshot. The
+	// live map keeps being mutated by subsequent rule operations while the
+	// state manager marshals the state from its periodic-save goroutine.
+	// Sharing it by reference races the two and aborts the process with a
+	// concurrent map iteration and write. The ipset counter guards itself
+	// during marshaling, so it can be shared directly.
 	if r.v6 {
-		currentState.RouteRules6 = r.rules
+		currentState.RouteRules6 = maps.Clone(r.rules)
 		currentState.RouteIPsetCounter6 = r.ipsetCounter
 	} else {
-		currentState.RouteRules = r.rules
+		currentState.RouteRules = maps.Clone(r.rules)
 		currentState.RouteIPsetCounter = r.ipsetCounter
 	}
 

client/firewall/iptables/router_linux_test.go

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

client/firewall/iptables/rulestore_linux.go

@@ -1,6 +1,9 @@
 package iptables
 
-import "encoding/json"
+import (
+	"encoding/json"
+	"maps"
+)
 
 type ipList struct {
 	ips map[string]struct{}
@@ -19,6 +22,14 @@ func (s *ipList) addIP(ip string) {
 	s.ips[ip] = struct{}{}
 }
 
+// clone returns a deep copy of the ipList with its own ips map.
+func (s *ipList) clone() *ipList {
+	if s == nil {
+		return nil
+	}
+	return &ipList{ips: maps.Clone(s.ips)}
+}
+
 // MarshalJSON implements json.Marshaler
 func (s *ipList) MarshalJSON() ([]byte, error) {
 	return json.Marshal(struct {
@@ -55,6 +66,19 @@ func newIpsetStore() *ipsetStore {
 	}
 }
 
+// clone returns a deep copy of the ipsetStore with its own ipsets map and
+// independent ipList entries.
+func (s *ipsetStore) clone() *ipsetStore {
+	if s == nil {
+		return nil
+	}
+	cloned := &ipsetStore{ipsets: make(map[string]*ipList, len(s.ipsets))}
+	for name, list := range s.ipsets {
+		cloned.ipsets[name] = list.clone()
+	}
+	return cloned
+}
+
 func (s *ipsetStore) ipset(ipsetName string) (*ipList, bool) {
 	r, ok := s.ipsets[ipsetName]
 	return r, ok

client/firewall/nftables/manager_linux_test.go

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

client/firewall/nftables/router_linux_test.go

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

client/iface/bind/ice_bind.go

@@ -41,7 +41,6 @@ type ICEBind struct {
 	*wgConn.StdNetBind
 
 	transportNet transport.Net
-	filterFn     udpmux.FilterFn
 	address      wgaddr.Address
 	mtu          uint16
 
@@ -61,12 +60,11 @@ type ICEBind struct {
 	ipv6Conn *net.UDPConn
 }
 
-func NewICEBind(transportNet transport.Net, filterFn udpmux.FilterFn, address wgaddr.Address, mtu uint16) *ICEBind {
+func NewICEBind(transportNet transport.Net, address wgaddr.Address, mtu uint16) *ICEBind {
 	b, _ := wgConn.NewStdNetBind().(*wgConn.StdNetBind)
 	ib := &ICEBind{
 		StdNetBind:       b,
 		transportNet:     transportNet,
-		filterFn:         filterFn,
 		address:          address,
 		mtu:              mtu,
 		endpoints:        make(map[netip.Addr]net.Conn),
@@ -265,7 +263,6 @@ func (s *ICEBind) createOrUpdateMux() {
 		udpmux.UniversalUDPMuxParams{
 			UDPConn:   muxConn,
 			Net:       s.transportNet,
-			FilterFn:  s.filterFn,
 			WGAddress: s.address,
 			MTU:       s.mtu,
 		},

client/iface/bind/ice_bind_test.go

@@ -289,7 +289,7 @@ func setupICEBind(t *testing.T) *ICEBind {
 		IP:      netip.MustParseAddr("100.64.0.1"),
 		Network: netip.MustParsePrefix("100.64.0.0/10"),
 	}
-	return NewICEBind(transportNet, nil, address, 1280)
+	return NewICEBind(transportNet, address, 1280)
 }
 
 func createDualStackConns(t *testing.T) (*net.UDPConn, *net.UDPConn) {

client/iface/device/device_filter.go

@@ -1,10 +1,13 @@
 package device
 
 import (
+	"fmt"
 	"net/netip"
+	"runtime/debug"
 	"sync"
 	"sync/atomic"
 
+	log "github.com/sirupsen/logrus"
 	"golang.zx2c4.com/wireguard/tun"
 )
 
@@ -41,10 +44,13 @@ type PacketCapture interface {
 type FilteredDevice struct {
 	tun.Device
 
-	filter    PacketFilter
+	filter  PacketFilter
-	capture   atomic.Pointer[PacketCapture]
+	capture atomic.Pointer[PacketCapture]
-	mutex     sync.RWMutex
+	// panicHandler is invoked after a panic in the underlying device is
-	closeOnce sync.Once
+	// recovered in Read or Write.
+	panicHandler atomic.Pointer[func()]
+	mutex        sync.RWMutex
+	closeOnce    sync.Once
 }
 
 // newDeviceFilter constructor function
@@ -70,7 +76,7 @@ func (d *FilteredDevice) Close() error {
 
 // Read wraps read method with filtering feature
 func (d *FilteredDevice) Read(bufs [][]byte, sizes []int, offset int) (n int, err error) {
-	if n, err = d.Device.Read(bufs, sizes, offset); err != nil {
+	if n, err = d.deviceRead(bufs, sizes, offset); err != nil {
 		return 0, err
 	}
 
@@ -112,7 +118,7 @@ func (d *FilteredDevice) Write(bufs [][]byte, offset int) (int, error) {
 	d.mutex.RUnlock()
 
 	if filter == nil {
-		return d.Device.Write(bufs, offset)
+		return d.deviceWrite(bufs, offset)
 	}
 
 	filteredBufs := make([][]byte, 0, len(bufs))
@@ -125,9 +131,44 @@ func (d *FilteredDevice) Write(bufs [][]byte, offset int) (int, error) {
 		}
 	}
 
-	n, err := d.Device.Write(filteredBufs, offset)
+	n, err := d.deviceWrite(filteredBufs, offset)
-	n += dropped
+	if err != nil {
-	return n, err
+		return n, err
+	}
+	return n + dropped, nil
+}
+
+// deviceRead calls the underlying device Read, recovering from panics in the
+// wintun read path and converting them into errors.
+func (d *FilteredDevice) deviceRead(bufs [][]byte, sizes []int, offset int) (n int, err error) {
+	defer d.recoverFromPanic("read", &n, &err)
+	return d.Device.Read(bufs, sizes, offset)
+}
+
+// deviceWrite calls the underlying device Write, recovering from panics in the
+// wintun write path and converting them into errors.
+func (d *FilteredDevice) deviceWrite(bufs [][]byte, offset int) (n int, err error) {
+	defer d.recoverFromPanic("write", &n, &err)
+	return d.Device.Write(bufs, offset)
+}
+
+// recoverFromPanic converts a panic in the underlying device into a regular
+// error and invokes the registered panic handler. The wintun read path is
+// known to panic on zero-length packets that third-party filter drivers can
+// place in the ring.
+func (d *FilteredDevice) recoverFromPanic(op string, n *int, err *error) {
+	r := recover()
+	if r == nil {
+		return
+	}
+
+	log.Errorf("recovered panic in tun device %s: %v\n%s", op, r, debug.Stack())
+	*n = 0
+	*err = fmt.Errorf("tun device %s panic: %v", op, r)
+
+	if handler := d.panicHandler.Load(); handler != nil {
+		(*handler)()
+	}
 }
 
 // SetFilter sets packet filter to device
@@ -137,6 +178,17 @@ func (d *FilteredDevice) SetFilter(filter PacketFilter) {
 	d.mutex.Unlock()
 }
 
+// SetPanicHandler registers a handler invoked after a recovered panic in Read
+// or Write. The device is unusable after such a panic; the handler should
+// trigger recreation of the interface. Pass nil to remove.
+func (d *FilteredDevice) SetPanicHandler(handler func()) {
+	if handler == nil {
+		d.panicHandler.Store(nil)
+		return
+	}
+	d.panicHandler.Store(&handler)
+}
+
 // SetCapture sets or clears the packet capture sink. Pass nil to disable.
 // Uses atomic store so the hot path (Read/Write) is a single pointer load
 // with no locking overhead when capture is off.

client/iface/device/device_filter_test.go

@@ -221,3 +221,60 @@ func TestDeviceWrapperRead(t *testing.T) {
 		}
 	})
 }
+
+func TestDeviceWrapperReadPanic(t *testing.T) {
+	ctrl := gomock.NewController(t)
+	defer ctrl.Finish()
+
+	tun := mocks.NewMockDevice(ctrl)
+	tun.EXPECT().Read(gomock.Any(), gomock.Any(), gomock.Any()).
+		DoAndReturn(func(bufs [][]byte, sizes []int, offset int) (int, error) {
+			// Reproduce the wintun zero-length packet panic (index out of range).
+			packet := make([]byte, 0)
+			return int(packet[0]), nil
+		})
+
+	wrapped := newDeviceFilter(tun)
+
+	handlerCalled := false
+	wrapped.SetPanicHandler(func() { handlerCalled = true })
+
+	n, err := wrapped.Read([][]byte{{}}, []int{0}, 0)
+	if err == nil {
+		t.Errorf("expected error from recovered panic, got nil")
+	}
+	if n != 0 {
+		t.Errorf("expected n=0, got %d", n)
+	}
+	if !handlerCalled {
+		t.Errorf("expected panic handler to be called")
+	}
+}
+
+func TestDeviceWrapperWritePanic(t *testing.T) {
+	ctrl := gomock.NewController(t)
+	defer ctrl.Finish()
+
+	tun := mocks.NewMockDevice(ctrl)
+	tun.EXPECT().Write(gomock.Any(), gomock.Any()).
+		DoAndReturn(func(bufs [][]byte, offset int) (int, error) {
+			packet := make([]byte, 0)
+			return int(packet[0]), nil
+		})
+
+	wrapped := newDeviceFilter(tun)
+
+	handlerCalled := false
+	wrapped.SetPanicHandler(func() { handlerCalled = true })
+
+	n, err := wrapped.Write([][]byte{{0x45, 0x00}}, 0)
+	if err == nil {
+		t.Errorf("expected error from recovered panic, got nil")
+	}
+	if n != 0 {
+		t.Errorf("expected n=0, got %d", n)
+	}
+	if !handlerCalled {
+		t.Errorf("expected panic handler to be called")
+	}
+}

client/iface/device/device_kernel_unix.go

@@ -32,8 +32,6 @@ type TunKernelDevice struct {
 	link       *wgLink
 	udpMuxConn net.PacketConn
 	udpMux     *udpmux.UniversalUDPMuxDefault
-
-	filterFn udpmux.FilterFn
 }
 
 func NewKernelDevice(name string, address wgaddr.Address, wgPort int, key string, mtu uint16, transportNet transport.Net) *TunKernelDevice {
@@ -104,7 +102,6 @@ func (t *TunKernelDevice) Up() (*udpmux.UniversalUDPMuxDefault, error) {
 	bindParams := udpmux.UniversalUDPMuxParams{
 		UDPConn:   nbnet.WrapPacketConn(rawSock),
 		Net:       t.transportNet,
-		FilterFn:  t.filterFn,
 		WGAddress: t.address,
 		MTU:       t.mtu,
 	}

client/iface/iface.go

@@ -63,7 +63,6 @@ type WGIFaceOpts struct {
 	MTU          uint16
 	MobileArgs   *device.MobileIFaceArguments
 	TransportNet transport.Net
-	FilterFn     udpmux.FilterFn
 	DisableDNS   bool
 }
 

client/iface/iface_new.go

@@ -11,7 +11,7 @@ import (
 
 // NewWGIFace Creates a new WireGuard interface instance
 func NewWGIFace(opts WGIFaceOpts) (*WGIface, error) {
-	iceBind := bind.NewICEBind(opts.TransportNet, opts.FilterFn, opts.Address, opts.MTU)
+	iceBind := bind.NewICEBind(opts.TransportNet, opts.Address, opts.MTU)
 
 	var tun WGTunDevice
 	if netstack.IsEnabled() {

client/iface/iface_new_android.go

@@ -9,7 +9,7 @@ import (
 
 // NewWGIFace Creates a new WireGuard interface instance
 func NewWGIFace(opts WGIFaceOpts) (*WGIface, error) {
-	iceBind := bind.NewICEBind(opts.TransportNet, opts.FilterFn, opts.Address, opts.MTU)
+	iceBind := bind.NewICEBind(opts.TransportNet, opts.Address, opts.MTU)
 
 	if netstack.IsEnabled() {
 		wgIFace := &WGIface{

client/iface/iface_new_ios.go

@@ -10,7 +10,7 @@ import (
 
 // NewWGIFace Creates a new WireGuard interface instance
 func NewWGIFace(opts WGIFaceOpts) (*WGIface, error) {
-	iceBind := bind.NewICEBind(opts.TransportNet, opts.FilterFn, opts.Address, opts.MTU)
+	iceBind := bind.NewICEBind(opts.TransportNet, opts.Address, opts.MTU)
 
 	wgIFace := &WGIface{
 		tun:            device.NewTunDevice(opts.IFaceName, opts.Address, opts.WGPort, opts.WGPrivKey, opts.MTU, iceBind, opts.MobileArgs.TunFd),

client/iface/iface_new_linux.go

@@ -14,7 +14,7 @@ import (
 // NewWGIFace Creates a new WireGuard interface instance
 func NewWGIFace(opts WGIFaceOpts) (*WGIface, error) {
 	if netstack.IsEnabled() {
-		iceBind := bind.NewICEBind(opts.TransportNet, opts.FilterFn, opts.Address, opts.MTU)
+		iceBind := bind.NewICEBind(opts.TransportNet, opts.Address, opts.MTU)
 		return &WGIface{
 			tun:            device.NewNetstackDevice(opts.IFaceName, opts.Address, opts.WGPort, opts.WGPrivKey, opts.MTU, iceBind, netstack.ListenAddr()),
 			userspaceBind:  true,
@@ -30,7 +30,7 @@ func NewWGIFace(opts WGIFaceOpts) (*WGIface, error) {
 	}
 
 	if device.ModuleTunIsLoaded() {
-		iceBind := bind.NewICEBind(opts.TransportNet, opts.FilterFn, opts.Address, opts.MTU)
+		iceBind := bind.NewICEBind(opts.TransportNet, opts.Address, opts.MTU)
 		return &WGIface{
 			tun:            device.NewTunDevice(opts.IFaceName, opts.Address, opts.WGPort, opts.WGPrivKey, opts.MTU, iceBind),
 			userspaceBind:  true,

client/iface/iface_test.go

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

client/iface/udpmux/universal.go

@@ -8,8 +8,6 @@ import (
 	"context"
 	"fmt"
 	"net"
-	"net/netip"
-	"sync"
 	"time"
 
 	log "github.com/sirupsen/logrus"
@@ -22,10 +20,6 @@ import (
 	"github.com/netbirdio/netbird/client/iface/wgaddr"
 )
 
-// FilterFn is a function that filters out candidates based on the address.
-// If it returns true, the address is to be filtered. It also returns the prefix of matching route.
-type FilterFn func(address netip.Addr) (bool, netip.Prefix, error)
-
 // UniversalUDPMuxDefault handles STUN and TURN servers packets by wrapping the original UDPConn
 // It then passes packets to the UDPMux that does the actual connection muxing.
 type UniversalUDPMuxDefault struct {
@@ -43,7 +37,6 @@ type UniversalUDPMuxParams struct {
 	UDPConn               net.PacketConn
 	XORMappedAddrCacheTTL time.Duration
 	Net                   transport.Net
-	FilterFn              FilterFn
 	WGAddress             wgaddr.Address
 	MTU                   uint16
 }
@@ -68,7 +61,6 @@ func NewUniversalUDPMuxDefault(params UniversalUDPMuxParams) *UniversalUDPMuxDef
 		PacketConn: params.UDPConn,
 		mux:        m,
 		logger:     params.Logger,
-		filterFn:   params.FilterFn,
 		address:    params.WGAddress,
 	}
 
@@ -115,15 +107,12 @@ func (m *UniversalUDPMuxDefault) ReadFromConn(ctx context.Context) {
 	}
 }
 
-// UDPConn is a wrapper around UDPMux conn that overrides ReadFrom and handles STUN/TURN packets
+// UDPConn is a wrapper around UDPMux conn that overrides WriteTo to drop packets destined for the overlay subnet.
 type UDPConn struct {
 	net.PacketConn
-	mux      *UniversalUDPMuxDefault
+	mux     *UniversalUDPMuxDefault
-	logger   logging.LeveledLogger
+	logger  logging.LeveledLogger
-	filterFn FilterFn
+	address wgaddr.Address
-	// TODO: reset cache on route changes
-	addrCache sync.Map
-	address   wgaddr.Address
 }
 
 // GetPacketConn returns the underlying PacketConn
@@ -132,65 +121,16 @@ func (u *UDPConn) GetPacketConn() net.PacketConn {
 }
 
 func (u *UDPConn) WriteTo(b []byte, addr net.Addr) (int, error) {
-	if u.filterFn == nil {
+	udpAddr, ok := addr.(*net.UDPAddr)
+	if !ok {
 		return u.PacketConn.WriteTo(b, addr)
 	}
-
+	dst := udpAddr.AddrPort().Addr().Unmap()
-	if isRouted, found := u.addrCache.Load(addr.String()); found {
+	if (u.address.Network.IsValid() && u.address.Network.Contains(dst)) || (u.address.IPv6Net.IsValid() && u.address.IPv6Net.Contains(dst)) {
-		return u.handleCachedAddress(isRouted.(bool), b, addr)
-	}
-
-	return u.handleUncachedAddress(b, addr)
-}
-
-func (u *UDPConn) handleCachedAddress(isRouted bool, b []byte, addr net.Addr) (int, error) {
-	if isRouted {
-		return 0, fmt.Errorf("address %s is part of a routed network, refusing to write", addr)
-	}
-	return u.PacketConn.WriteTo(b, addr)
-}
-
-func (u *UDPConn) handleUncachedAddress(b []byte, addr net.Addr) (int, error) {
-	if err := u.performFilterCheck(addr); err != nil {
-		return 0, err
-	}
-	return u.PacketConn.WriteTo(b, addr)
-}
-
-func (u *UDPConn) performFilterCheck(addr net.Addr) error {
-	host, err := getHostFromAddr(addr)
-	if err != nil {
-		log.Errorf("Failed to get host from address %s: %v", addr, err)
-		return nil
-	}
-
-	a, err := netip.ParseAddr(host)
-	if err != nil {
-		log.Errorf("Failed to parse address %s: %v", addr, err)
-		return nil
-	}
-
-	if u.address.Network.Contains(a) {
 		log.Warnf("address %s is part of the NetBird network %s, refusing to write", addr, u.address)
-		return fmt.Errorf("address %s is part of the NetBird network %s, refusing to write", addr, u.address)
+		return 0, fmt.Errorf("address %s is part of the NetBird network %s, refusing to write", addr, u.address)
 	}
-
+	return u.PacketConn.WriteTo(b, addr)
-	if isRouted, prefix, err := u.filterFn(a); err != nil {
-		log.Errorf("Failed to check if address %s is routed: %v", addr, err)
-	} else {
-		u.addrCache.Store(addr.String(), isRouted)
-		if isRouted {
-			// Extra log, as the error only shows up with ICE logging enabled
-			log.Infof("address %s is part of routed network %s, refusing to write", addr, prefix)
-			return fmt.Errorf("address %s is part of routed network %s, refusing to write", addr, prefix)
-		}
-	}
-	return nil
-}
-
-func getHostFromAddr(addr net.Addr) (string, error) {
-	host, _, err := net.SplitHostPort(addr.String())
-	return host, err
 }
 
 // GetSharedConn returns the shared udp conn
@@ -225,6 +165,13 @@ func (m *UniversalUDPMuxDefault) HandleSTUNMessage(msg *stun.Message, addr net.A
 		return nil
 	}
 
+	src := udpAddr.AddrPort().Addr().Unmap()
+	wg := m.params.WGAddress
+	if (wg.Network.IsValid() && wg.Network.Contains(src)) || (wg.IPv6Net.IsValid() && wg.IPv6Net.Contains(src)) {
+		log.Debugf("dropping STUN message from overlay source %s", udpAddr)
+		return nil
+	}
+
 	if m.isXORMappedResponse(msg, udpAddr.String()) {
 		err := m.handleXORMappedResponse(udpAddr, msg)
 		if err != nil {

client/iface/wgproxy/proxy_linux_test.go

@@ -1,4 +1,4 @@
-//go:build linux && !android
+//go:build linux && !android && privileged
 
 package wgproxy
 
@@ -66,7 +66,7 @@ func seedProxyForProxyCloseByRemoteConn() ([]proxyInstance, error) {
 	if err != nil {
 		return nil, err
 	}
-	iceBind := bind.NewICEBind(nil, nil, wgAddress, 1280)
+	iceBind := bind.NewICEBind(nil, wgAddress, 1280)
 	endpointAddress := &net.UDPAddr{
 		IP:   net.IPv4(10, 0, 0, 1),
 		Port: 1234,

client/iface/wgproxy/proxy_seed_test.go

@@ -1,4 +1,4 @@
-//go:build !linux
+//go:build !linux || !privileged
 
 package wgproxy
 
@@ -22,7 +22,7 @@ func seedProxyForProxyCloseByRemoteConn() ([]proxyInstance, error) {
 	if err != nil {
 		return nil, err
 	}
-	iceBind := bind.NewICEBind(nil, nil, wgAddress, 1280)
+	iceBind := bind.NewICEBind(nil, wgAddress, 1280)
 	endpointAddress := &net.UDPAddr{
 		IP:   net.IPv4(10, 0, 0, 1),
 		Port: 1234,

client/iface/wgproxy/redirect_test.go

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

client/internal/connect.go

@@ -11,6 +11,7 @@ import (
 	"runtime/debug"
 	"strings"
 	"sync"
+	"sync/atomic"
 	"time"
 
 	"github.com/cenkalti/backoff/v4"
@@ -54,6 +55,10 @@ var androidRunOverride func(c *ConnectClient, runningChan chan struct{}, logPath
 
 type ConnectClient struct {
 	ctx            context.Context
+	runCancel      context.CancelFunc
+	runExited      chan struct{}
+	runOnce        sync.Once
+	runStarted     atomic.Bool
 	config         *profilemanager.Config
 	statusRecorder *peer.Status
 
@@ -70,8 +75,14 @@ func NewConnectClient(
 	config *profilemanager.Config,
 	statusRecorder *peer.Status,
 ) *ConnectClient {
+	// Derive the run context here so Stop owns the cancel that unblocks the run
+	// loop. runCancel is set once at construction, so Stop can call it without
+	// racing the run loop's startup. Callers therefore need not cancel before Stop.
+	runCtx, runCancel := context.WithCancel(ctx)
 	return &ConnectClient{
-		ctx:            ctx,
+		ctx:            runCtx,
+		runCancel:      runCancel,
+		runExited:      make(chan struct{}),
 		config:         config,
 		statusRecorder: statusRecorder,
 		engineMutex:    sync.Mutex{},
@@ -118,6 +129,8 @@ func (c *ConnectClient) RunOniOS(
 	networkChangeListener listener.NetworkChangeListener,
 	dnsManager dns.IosDnsManager,
 	stateFilePath string,
+	cacheDir string,
+	logFilePath string,
 ) error {
 	// Set GC percent to 5% to reduce memory usage as iOS only allows 50MB of memory for the extension.
 	debug.SetGCPercent(5)
@@ -127,11 +140,17 @@ func (c *ConnectClient) RunOniOS(
 		NetworkChangeListener: networkChangeListener,
 		DnsManager:            dnsManager,
 		StateFilePath:         stateFilePath,
+		TempDir:               cacheDir,
 	}
-	return c.run(mobileDependency, nil, "")
+	return c.run(mobileDependency, nil, logFilePath)
 }
 
 func (c *ConnectClient) run(mobileDependency MobileDependency, runningChan chan struct{}, logPath string) error {
+	// Mark the loop as started and signal exit on return so Stop can wait for
+	// the loop to finish (and skip the wait if the loop never ran).
+	c.runStarted.Store(true)
+	defer c.runOnce.Do(func() { close(c.runExited) })
+
 	defer func() {
 		if r := recover(); r != nil {
 			rec := c.statusRecorder
@@ -287,7 +306,7 @@ func (c *ConnectClient) run(mobileDependency MobileDependency, runningChan chan
 			log.Debug(err)
 			if s, ok := gstatus.FromError(err); ok && (s.Code() == codes.PermissionDenied) {
 				state.Set(StatusNeedsLogin)
-				_ = c.Stop()
+				c.runCancel()
 				return backoff.Permanent(wrapErr(err)) // unrecoverable error
 			}
 			return wrapErr(err)
@@ -407,14 +426,10 @@ func (c *ConnectClient) run(mobileDependency MobileDependency, runningChan chan
 		c.engine = nil
 		c.engineMutex.Unlock()
 
-		// todo: consider to remove this condition. Is not thread safe.
+		log.Infof("ensuring wg interface is removed, Netbird engine context cancelled")
-		// We should always call Stop(), but we need to verify that it is idempotent
-		if engine.wgInterface != nil {
-			log.Infof("ensuring %s is removed, Netbird engine context cancelled", engine.wgInterface.Name())
 
-			if err := engine.Stop(); err != nil {
+		if err := engine.Stop(); err != nil {
-				log.Errorf("Failed to stop engine: %v", err)
+			log.Errorf("Failed to stop engine: %v", err)
-			}
 		}
 		c.statusRecorder.ClientTeardown()
 
@@ -430,12 +445,12 @@ func (c *ConnectClient) run(mobileDependency MobileDependency, runningChan chan
 	}
 
 	c.statusRecorder.ClientStart()
-	err = backoff.Retry(operation, backOff)
+	err = backoff.Retry(operation, backoff.WithContext(backOff, c.ctx))
 	if err != nil {
 		log.Debugf("exiting client retry loop due to unrecoverable error: %s", err)
 		if s, ok := gstatus.FromError(err); ok && (s.Code() == codes.PermissionDenied) {
 			state.Set(StatusNeedsLogin)
-			_ = c.Stop()
+			c.runCancel()
 		}
 		return err
 	}
@@ -513,11 +528,9 @@ func (c *ConnectClient) Status() StatusType {
 }
 
 func (c *ConnectClient) Stop() error {
-	engine := c.Engine()
+	c.runCancel()
-	if engine != nil {
+	if c.runStarted.Load() {
-		if err := engine.Stop(); err != nil {
+		<-c.runExited
-			return fmt.Errorf("stop engine: %w", err)
-		}
 	}
 	return nil
 }

client/internal/debug/debug.go

@@ -250,6 +250,7 @@ type BundleGenerator struct {
 	syncResponse   *mgmProto.SyncResponse
 	logPath        string
 	tempDir        string
+	statePath      string
 	cpuProfile     []byte
 	capturePath    string
 	refreshStatus  func() // Optional callback to refresh status before bundle generation
@@ -276,6 +277,7 @@ type GeneratorDependencies struct {
 	SyncResponse   *mgmProto.SyncResponse
 	LogPath        string
 	TempDir        string // Directory for temporary bundle zip files. If empty, os.TempDir() is used.
+	StatePath      string // Path to the state file. If empty, the ServiceManager default path is used.
 	CPUProfile     []byte
 	CapturePath    string
 	RefreshStatus  func()
@@ -299,6 +301,7 @@ func NewBundleGenerator(deps GeneratorDependencies, cfg BundleConfig) *BundleGen
 		syncResponse:   deps.SyncResponse,
 		logPath:        deps.LogPath,
 		tempDir:        deps.TempDir,
+		statePath:      deps.StatePath,
 		cpuProfile:     deps.CPUProfile,
 		capturePath:    deps.CapturePath,
 		refreshStatus:  deps.RefreshStatus,
@@ -516,6 +519,14 @@ func (g *BundleGenerator) addConfig() error {
 		}
 	}
 
+	// Surface the set of MDM-enforced keys so a support engineer reading
+	// the bundle can tell which field values are user-set vs MDM-overridden.
+	// Same semantics as the mDMManagedFields list returned by the
+	// GetConfig RPC consumed by `netbird debug config`.
+	if managed := g.internalConfig.Policy().ManagedKeys(); len(managed) > 0 {
+		configContent.WriteString(fmt.Sprintf("MDMManagedFields: %v\n", managed))
+	}
+
 	configReader := strings.NewReader(configContent.String())
 	if err := g.addFileToZip(configReader, "config.txt"); err != nil {
 		return fmt.Errorf("add config file to zip: %w", err)
@@ -842,8 +853,11 @@ func (g *BundleGenerator) maskSecrets() {
 }
 
 func (g *BundleGenerator) addStateFile() error {
-	sm := profilemanager.NewServiceManager("")
+	path := g.statePath
-	path := sm.GetStatePath()
+	if path == "" {
+		sm := profilemanager.NewServiceManager("")
+		path = sm.GetStatePath()
+	}
 	if path == "" {
 		return nil
 	}

client/internal/debug/debug_ios.go

@@ -0,0 +1,36 @@
+//go:build ios
+
+package debug
+
+import (
+	"path/filepath"
+
+	log "github.com/sirupsen/logrus"
+)
+
+// swiftLogFile is the Swift app log written by the iOS app into the same log
+// directory as the Go client log, so it can be collected into the bundle.
+const swiftLogFile = "swift-log.log"
+
+// addPlatformLog collects logs for the iOS debug bundle. iOS has no logcat or
+// systemd journal, so we rely on file-based logs. addLogfile handles the Go
+// client log (logPath) with rotation, the stderr/stdout companions and
+// anonymization. The iOS app writes its own Swift log into the same directory,
+// so we add it alongside the Go log.
+func (g *BundleGenerator) addPlatformLog() error {
+	if err := g.addLogfile(); err != nil {
+		return err
+	}
+
+	if g.logPath == "" {
+		return nil
+	}
+
+	swiftLogPath := filepath.Join(filepath.Dir(g.logPath), swiftLogFile)
+	if err := g.addSingleLogfile(swiftLogPath, swiftLogFile); err != nil {
+		// The Swift log is best-effort: the app may not have written it yet.
+		log.Warnf("failed to add %s to debug bundle: %v", swiftLogFile, err)
+	}
+
+	return nil
+}

client/internal/debug/debug_nonandroid.go

@@ -1,4 +1,4 @@
-//go:build !android
+//go:build !android && !ios
 
 package debug
 

client/internal/debug/debug_test.go

@@ -843,6 +843,8 @@ func TestAddConfig_AllFieldsCovered(t *testing.T) {
 		"PreSharedKey":      "sensitive: WireGuard pre-shared key",
 		"SSHKey":            "sensitive: SSH private key",
 		"ClientCertKeyPair": "non-config: parsed cert pair, not serialized",
+		"Name":              "non-config: profile name is not needed for debug purposes",
+		"policy":            "non-config: in-memory MDM policy snapshot, surfaced via Config.Policy() / GetConfigResponse.MDMManagedFields",
 	}
 
 	mURL, _ := url.Parse("https://api.example.com:443")

client/internal/dns/local/local.go

@@ -482,7 +482,7 @@ func (d *Resolver) logDNSError(logger *log.Entry, hostname string, qtype uint16,
 // completely when every proxy peer is offline (the upstream may still
 // be reachable some other way, or the peerstore may be stale).
 func (d *Resolver) filterDisconnectedPeerAnswers(logger *log.Entry, question dns.Question, records []dns.RR) []dns.RR {
-	if len(records) == 0 {
+	if len(records) < 2 {
 		return records
 	}
 	d.mu.RLock()

client/internal/dns/local/local_test.go

@@ -2738,6 +2738,17 @@ func TestLocalResolver_FilterDisconnectedPeerAnswers(t *testing.T) {
 			connByIP:    nil,
 			wantInOrder: []string{"100.64.0.10", "100.64.0.11"},
 		},
+		{
+			// A single answer is never filtered: dropping it would only
+			// trigger the empty-answer escape hatch, so the fast path
+			// returns it untouched.
+			name:    "single disconnected answer passes through",
+			records: []nbdns.SimpleRecord{disconnectedRec},
+			connByIP: map[string]ipState{
+				"100.64.0.11": {known: true, connected: false},
+			},
+			wantInOrder: []string{"100.64.0.11"},
+		},
 	}
 
 	for _, tc := range tests {

client/internal/dns/mgmt/mgmt.go

@@ -51,13 +51,20 @@ type cachedRecord struct {
 }
 
 // Resolver caches critical NetBird infrastructure domains.
-// records, refreshing, mgmtDomain and serverDomains are all guarded by mutex.
+// records, refreshing, failedResolves, mgmtDomain and serverDomains are all
+// guarded by mutex.
 type Resolver struct {
 	records       map[dns.Question]*cachedRecord
 	mgmtDomain    *domain.Domain
 	serverDomains *dnsconfig.ServerDomains
 	mutex         sync.RWMutex
 
+	// failedResolves records the last failed initial resolve per domain so a
+	// domain that never resolves isn't retried on every server-domains update
+	// until refreshBackoff elapses. Entries are cleared on success and pruned
+	// to the current server-domains set.
+	failedResolves map[domain.Domain]time.Time
+
 	chain            ChainResolver
 	chainMaxPriority int
 	refreshGroup     singleflight.Group
@@ -76,9 +83,10 @@ type Resolver struct {
 // NewResolver creates a new management domains cache resolver.
 func NewResolver() *Resolver {
 	return &Resolver{
-		records:    make(map[dns.Question]*cachedRecord),
+		records:        make(map[dns.Question]*cachedRecord),
-		refreshing: make(map[dns.Question]*atomic.Bool),
+		refreshing:     make(map[dns.Question]*atomic.Bool),
-		cacheTTL:   resolveCacheTTL(),
+		failedResolves: make(map[domain.Domain]time.Time),
+		cacheTTL:       resolveCacheTTL(),
 	}
 }
 
@@ -173,7 +181,9 @@ func (m *Resolver) continueToNext(w dns.ResponseWriter, r *dns.Msg) {
 
 // AddDomain resolves a domain and stores its A/AAAA records in the cache.
 // A family that resolves NODATA (nil err, zero records) evicts any stale
-// entry for that qtype.
+// entry for that qtype. When one family hard-errors while the other succeeds,
+// the resolved family is still cached but AddDomain returns an error so the
+// caller retries the incomplete resolve rather than treating it as complete.
 func (m *Resolver) AddDomain(ctx context.Context, d domain.Domain) error {
 	dnsName := strings.ToLower(dns.Fqdn(d.PunycodeString()))
 
@@ -203,6 +213,10 @@ func (m *Resolver) AddDomain(ctx context.Context, d domain.Domain) error {
 	log.Debugf("added/updated domain=%s with %d A records and %d AAAA records",
 		d.SafeString(), len(aRecords), len(aaaaRecords))
 
+	if errA != nil || errAAAA != nil {
+		return fmt.Errorf("resolve %s: incomplete, a family failed: %w", d.SafeString(), errors.Join(errA, errAAAA))
+	}
+
 	return nil
 }
 
@@ -462,6 +476,7 @@ func (m *Resolver) RemoveDomain(d domain.Domain) error {
 	delete(m.records, qAAAA)
 	delete(m.refreshing, qA)
 	delete(m.refreshing, qAAAA)
+	delete(m.failedResolves, d)
 
 	log.Debugf("removed domain=%s from cache", d.SafeString())
 	return nil
@@ -505,6 +520,7 @@ func (m *Resolver) UpdateFromServerDomains(ctx context.Context, serverDomains dn
 		allDomains := m.extractDomainsFromServerDomains(updatedServerDomains)
 		currentDomains := m.GetCachedDomains()
 		removedDomains = m.removeStaleDomains(currentDomains, allDomains)
+		m.pruneFailedResolves(allDomains)
 	}
 
 	m.addNewDomains(ctx, newDomains)
@@ -577,13 +593,85 @@ func (m *Resolver) isManagementDomain(domain domain.Domain) bool {
 	return m.mgmtDomain != nil && domain == *m.mgmtDomain
 }
 
-// addNewDomains resolves and caches all domains from the update
+// addNewDomains resolves and caches domains that are not yet in the cache,
+// running the lookups concurrently. Domains already cached are skipped and left
+// to the stale-while-revalidate refresh path, so a sync never re-resolves them
+// synchronously: once NetBird owns the OS resolver the resolve runs through the
+// handler chain and would otherwise dial the managed upstreams under the engine
+// sync lock on every update.
 func (m *Resolver) addNewDomains(ctx context.Context, newDomains domain.List) {
+	var wg sync.WaitGroup
+	seen := make(map[domain.Domain]struct{}, len(newDomains))
 	for _, newDomain := range newDomains {
-		if err := m.AddDomain(ctx, newDomain); err != nil {
+		if _, dup := seen[newDomain]; dup {
-			log.Warnf("failed to add/update domain=%s: %v", newDomain.SafeString(), err)
+			continue
-		} else {
+		}
-			log.Debugf("added/updated management cache domain=%s", newDomain.SafeString())
+		seen[newDomain] = struct{}{}
+
+		if !m.needsResolve(newDomain) {
+			continue
+		}
+
+		wg.Add(1)
+		go func(d domain.Domain) {
+			defer wg.Done()
+			if err := m.AddDomain(ctx, d); err != nil {
+				m.markResolveFailed(d)
+				log.Warnf("failed to add/update domain=%s: %v", d.SafeString(), err)
+				return
+			}
+			m.clearResolveFailed(d)
+			log.Debugf("added/updated management cache domain=%s", d.SafeString())
+		}(newDomain)
+	}
+	wg.Wait()
+}
+
+// needsResolve reports whether d should be resolved now. A recent failed or
+// incomplete resolve gates retries on the backoff even when one family is
+// already cached, so a transiently-failed family is retried instead of being
+// treated as fully resolved. Otherwise a domain with any cached record is left
+// to the stale-while-revalidate refresh path.
+func (m *Resolver) needsResolve(d domain.Domain) bool {
+	dnsName := strings.ToLower(dns.Fqdn(d.PunycodeString()))
+
+	m.mutex.RLock()
+	defer m.mutex.RUnlock()
+
+	if failedAt, ok := m.failedResolves[d]; ok {
+		return time.Since(failedAt) >= refreshBackoff
+	}
+
+	for _, qtype := range []uint16{dns.TypeA, dns.TypeAAAA} {
+		q := dns.Question{Name: dnsName, Qtype: qtype, Qclass: dns.ClassINET}
+		if _, ok := m.records[q]; ok {
+			return false
+		}
+	}
+	return true
+}
+
+func (m *Resolver) markResolveFailed(d domain.Domain) {
+	m.mutex.Lock()
+	m.failedResolves[d] = time.Now()
+	m.mutex.Unlock()
+}
+
+func (m *Resolver) clearResolveFailed(d domain.Domain) {
+	m.mutex.Lock()
+	delete(m.failedResolves, d)
+	m.mutex.Unlock()
+}
+
+// pruneFailedResolves drops failure markers for domains no longer present in
+// the server-domains set, keeping the map bounded to the current set (a
+// failed-only domain has no cached record, so RemoveDomain never sees it).
+func (m *Resolver) pruneFailedResolves(domains domain.List) {
+	m.mutex.Lock()
+	defer m.mutex.Unlock()
+	for d := range m.failedResolves {
+		if !slices.Contains(domains, d) {
+			delete(m.failedResolves, d)
 		}
 	}
 }

client/internal/dns/mgmt/mgmt_refresh_test.go

@@ -21,6 +21,7 @@ type fakeChain struct {
 	mu       sync.Mutex
 	calls    map[string]int
 	answers  map[string][]dns.RR
+	qErr     map[string]error
 	err      error
 	hasRoot  bool
 	onLookup func()
@@ -30,6 +31,7 @@ func newFakeChain() *fakeChain {
 	return &fakeChain{
 		calls:   map[string]int{},
 		answers: map[string][]dns.RR{},
+		qErr:    map[string]error{},
 		hasRoot: true,
 	}
 }
@@ -47,6 +49,9 @@ func (f *fakeChain) ResolveInternal(ctx context.Context, msg *dns.Msg, maxPriori
 	f.calls[key]++
 	answers := f.answers[key]
 	err := f.err
+	if err == nil {
+		err = f.qErr[key]
+	}
 	onLookup := f.onLookup
 	f.mu.Unlock()
 
@@ -75,6 +80,12 @@ func (f *fakeChain) setAnswer(name string, qtype uint16, ip string) {
 	}
 }
 
+func (f *fakeChain) setErr(name string, qtype uint16, err error) {
+	f.mu.Lock()
+	defer f.mu.Unlock()
+	f.qErr[name+"|"+dns.TypeToString[qtype]] = err
+}
+
 func (f *fakeChain) callCount(name string, qtype uint16) int {
 	f.mu.Lock()
 	defer f.mu.Unlock()

client/internal/dns/mgmt/mgmt_resolve_test.go

@@ -0,0 +1,183 @@
+package mgmt
+
+import (
+	"context"
+	"errors"
+	"sync/atomic"
+	"testing"
+	"time"
+
+	"github.com/miekg/dns"
+	"github.com/stretchr/testify/assert"
+	"github.com/stretchr/testify/require"
+
+	dnsconfig "github.com/netbirdio/netbird/client/internal/dns/config"
+	"github.com/netbirdio/netbird/shared/management/domain"
+)
+
+// A domain already in the cache must not be re-resolved on a subsequent server
+// domains update; it is left to the stale-while-revalidate refresh path.
+func TestResolver_UpdateFromServerDomains_SkipsCached(t *testing.T) {
+	r := NewResolver()
+	chain := newFakeChain()
+	chain.setAnswer("signal.example.com.", dns.TypeA, "10.0.0.2")
+	r.SetChainResolver(chain, 50)
+
+	sd := dnsconfig.ServerDomains{Signal: domain.Domain("signal.example.com")}
+
+	_, err := r.UpdateFromServerDomains(context.Background(), sd)
+	require.NoError(t, err)
+	require.Equal(t, 1, chain.callCount("signal.example.com.", dns.TypeA),
+		"first update must resolve the domain")
+
+	_, err = r.UpdateFromServerDomains(context.Background(), sd)
+	require.NoError(t, err)
+	assert.Equal(t, 1, chain.callCount("signal.example.com.", dns.TypeA),
+		"cached domain must not be re-resolved on a subsequent update")
+}
+
+// New domains in a single update must resolve concurrently rather than serially.
+func TestResolver_AddNewDomains_ResolvesConcurrently(t *testing.T) {
+	r := NewResolver()
+	chain := newFakeChain()
+
+	var inflight, maxInflight atomic.Int32
+	chain.onLookup = func() {
+		n := inflight.Add(1)
+		for {
+			old := maxInflight.Load()
+			if n <= old || maxInflight.CompareAndSwap(old, n) {
+				break
+			}
+		}
+		time.Sleep(50 * time.Millisecond)
+		inflight.Add(-1)
+	}
+
+	relays := []domain.Domain{"a.example.com", "b.example.com", "c.example.com", "d.example.com"}
+	for _, d := range relays {
+		chain.setAnswer(dns.Fqdn(string(d)), dns.TypeA, "10.0.0.2")
+	}
+	r.SetChainResolver(chain, 50)
+
+	start := time.Now()
+	_, err := r.UpdateFromServerDomains(context.Background(), dnsconfig.ServerDomains{Relay: relays})
+	require.NoError(t, err)
+	elapsed := time.Since(start)
+
+	assert.GreaterOrEqual(t, int(maxInflight.Load()), 2, "domains must resolve concurrently")
+	// Serial resolution of 4 domains would take at least 4*50ms; concurrent is far less.
+	assert.Less(t, elapsed, 300*time.Millisecond, "resolution should not be serial")
+}
+
+// A domain that fails to resolve must not be retried on every update; the
+// failure backoff suppresses re-resolution until it expires.
+func TestResolver_UpdateFromServerDomains_BacksOffFailures(t *testing.T) {
+	r := NewResolver()
+	chain := newFakeChain()
+	chain.err = errors.New("resolve boom")
+	r.SetChainResolver(chain, 50)
+
+	sd := dnsconfig.ServerDomains{Signal: domain.Domain("signal.example.com")}
+
+	_, err := r.UpdateFromServerDomains(context.Background(), sd)
+	require.NoError(t, err)
+	require.Equal(t, 1, chain.callCount("signal.example.com.", dns.TypeA),
+		"first update must attempt the resolve")
+
+	_, err = r.UpdateFromServerDomains(context.Background(), sd)
+	require.NoError(t, err)
+	assert.Equal(t, 1, chain.callCount("signal.example.com.", dns.TypeA),
+		"failed resolve must back off and not retry on the next update")
+}
+
+// A domain listed under more than one server-domain type (e.g. STUN and TURN on
+// the same host) must be resolved once per update, not once per occurrence.
+func TestResolver_AddNewDomains_DedupesDuplicateDomains(t *testing.T) {
+	r := NewResolver()
+	chain := newFakeChain()
+	chain.setAnswer("dup.example.com.", dns.TypeA, "10.0.0.9")
+	r.SetChainResolver(chain, 50)
+
+	sd := dnsconfig.ServerDomains{
+		Stuns: []domain.Domain{"dup.example.com"},
+		Turns: []domain.Domain{"dup.example.com"},
+	}
+
+	_, err := r.UpdateFromServerDomains(context.Background(), sd)
+	require.NoError(t, err)
+	assert.Equal(t, 1, chain.callCount("dup.example.com.", dns.TypeA),
+		"a domain appearing under multiple server-domain types must resolve once")
+}
+
+// A failure marker must be dropped once its domain leaves the server-domains set
+// so the map stays bounded to the current set.
+func TestResolver_UpdateFromServerDomains_PrunesFailedResolves(t *testing.T) {
+	r := NewResolver()
+	chain := newFakeChain()
+	chain.err = errors.New("resolve boom")
+	r.SetChainResolver(chain, 50)
+
+	_, err := r.UpdateFromServerDomains(context.Background(), dnsconfig.ServerDomains{Signal: domain.Domain("gone.example.com")})
+	require.NoError(t, err)
+	r.mutex.RLock()
+	_, marked := r.failedResolves[domain.Domain("gone.example.com")]
+	r.mutex.RUnlock()
+	require.True(t, marked, "failed resolve must be recorded")
+
+	_, err = r.UpdateFromServerDomains(context.Background(), dnsconfig.ServerDomains{Signal: domain.Domain("other.example.com")})
+	require.NoError(t, err)
+	r.mutex.RLock()
+	_, stillMarked := r.failedResolves[domain.Domain("gone.example.com")]
+	r.mutex.RUnlock()
+	assert.False(t, stillMarked, "failure marker for a domain no longer in the set must be pruned")
+}
+
+// When one family hard-errors while the other resolves, the domain is cached
+// for the working family but recorded as incomplete so the failed family is
+// retried under backoff instead of being treated as fully resolved forever.
+func TestResolver_AddNewDomains_RetriesPartialFamilyFailure(t *testing.T) {
+	d := domain.Domain("relay.example.com")
+	r := NewResolver()
+	chain := newFakeChain()
+	chain.setAnswer("relay.example.com.", dns.TypeA, "10.0.0.2")
+	chain.setErr("relay.example.com.", dns.TypeAAAA, errors.New("servfail"))
+	r.SetChainResolver(chain, 50)
+
+	_, err := r.UpdateFromServerDomains(context.Background(), dnsconfig.ServerDomains{Relay: []domain.Domain{d}})
+	require.NoError(t, err)
+
+	r.mutex.RLock()
+	_, aCached := r.records[dns.Question{Name: "relay.example.com.", Qtype: dns.TypeA, Qclass: dns.ClassINET}]
+	_, marked := r.failedResolves[d]
+	r.mutex.RUnlock()
+	require.True(t, aCached, "the working family must still be cached")
+	require.True(t, marked, "a partial failure must be recorded so the failed family is retried")
+
+	assert.False(t, r.needsResolve(d), "within the backoff window the domain is not retried")
+
+	r.mutex.Lock()
+	r.failedResolves[d] = time.Now().Add(-2 * refreshBackoff)
+	r.mutex.Unlock()
+	assert.True(t, r.needsResolve(d), "after the backoff elapses the domain is retried to pick up the missing family")
+}
+
+// A family that returns NODATA (legitimately absent, e.g. an IPv4-only host) is
+// not a failure: the domain must not be marked for retry, otherwise it would be
+// re-resolved on every sync.
+func TestResolver_AddNewDomains_NodataIsNotFailure(t *testing.T) {
+	d := domain.Domain("v4only.example.com")
+	r := NewResolver()
+	chain := newFakeChain()
+	chain.setAnswer("v4only.example.com.", dns.TypeA, "10.0.0.2")
+	r.SetChainResolver(chain, 50)
+
+	_, err := r.UpdateFromServerDomains(context.Background(), dnsconfig.ServerDomains{Relay: []domain.Domain{d}})
+	require.NoError(t, err)
+
+	r.mutex.RLock()
+	_, marked := r.failedResolves[d]
+	r.mutex.RUnlock()
+	assert.False(t, marked, "a NODATA family must not be recorded as a failure")
+	assert.False(t, r.needsResolve(d), "an IPv4-only host must not be re-resolved on later syncs")
+}

client/internal/dns/resutil/resolve.go

@@ -14,6 +14,10 @@ import (
 	log "github.com/sirupsen/logrus"
 )
 
+// errNoSuitableAddress mirrors the unexported error string the net package
+// uses when a resolved host has no addresses of the requested family.
+const errNoSuitableAddress = "no suitable address found"
+
 // GenerateRequestID creates a random 8-character hex string for request tracing.
 func GenerateRequestID() string {
 	bytes := make([]byte, 4)
@@ -126,6 +130,14 @@ func LookupIP(ctx context.Context, r resolver, network, host string, qtype uint1
 }
 
 func getRcodeForError(ctx context.Context, r resolver, host string, qtype uint16, err error) int {
+	// The net package returns this AddrError when the host resolves but has
+	// no addresses of the requested family. The domain exists, so answer
+	// NODATA instead of SERVFAIL.
+	var addrErr *net.AddrError
+	if errors.As(err, &addrErr) && addrErr.Err == errNoSuitableAddress {
+		return dns.RcodeSuccess
+	}
+
 	var dnsErr *net.DNSError
 	if !errors.As(err, &dnsErr) {
 		return dns.RcodeServerFailure
@@ -195,3 +207,35 @@ func FormatAnswers(answers []dns.RR) string {
 	}
 	return "[" + strings.Join(parts, ", ") + "]"
 }
+
+// StripOPT removes any OPT pseudo-RRs from the message's Extra section. Per
+// RFC 6891 a responder must not include an OPT RR toward a client that did not
+// advertise EDNS0.
+func StripOPT(msg *dns.Msg) {
+	if len(msg.Extra) == 0 {
+		return
+	}
+	out := msg.Extra[:0]
+	for _, rr := range msg.Extra {
+		if _, ok := rr.(*dns.OPT); ok {
+			continue
+		}
+		out = append(out, rr)
+	}
+	msg.Extra = out
+}
+
+// ExtractEDE returns the first Extended DNS Error (RFC 8914) option carried in
+// the message, if present.
+func ExtractEDE(msg *dns.Msg) (*dns.EDNS0_EDE, bool) {
+	opt := msg.IsEdns0()
+	if opt == nil {
+		return nil, false
+	}
+	for _, o := range opt.Option {
+		if ede, ok := o.(*dns.EDNS0_EDE); ok {
+			return ede, true
+		}
+	}
+	return nil, false
+}

client/internal/dns/resutil/resolve_test.go

@@ -0,0 +1,161 @@
+package resutil
+
+import (
+	"context"
+	"errors"
+	"net"
+	"net/netip"
+	"testing"
+
+	"github.com/miekg/dns"
+	"github.com/stretchr/testify/assert"
+	"github.com/stretchr/testify/require"
+)
+
+type mockResolver struct {
+	// results maps network ("ip4"/"ip6") to the lookup outcome.
+	results map[string]mockLookup
+}
+
+type mockLookup struct {
+	ips []netip.Addr
+	err error
+}
+
+func (m *mockResolver) LookupNetIP(_ context.Context, network, _ string) ([]netip.Addr, error) {
+	res, ok := m.results[network]
+	if !ok {
+		return nil, errors.New("unexpected network: " + network)
+	}
+	return res.ips, res.err
+}
+
+func TestLookupIP_Success(t *testing.T) {
+	r := &mockResolver{results: map[string]mockLookup{
+		"ip4": {ips: []netip.Addr{netip.MustParseAddr("::ffff:192.0.2.1")}},
+	}}
+
+	result := LookupIP(context.Background(), r, "ip4", "example.com.", dns.TypeA)
+
+	assert.Equal(t, dns.RcodeSuccess, result.Rcode, "successful lookup should return NOERROR")
+	require.Len(t, result.IPs, 1, "should return the resolved address")
+	assert.Equal(t, netip.MustParseAddr("192.0.2.1"), result.IPs[0], "v4-mapped address should be unmapped")
+}
+
+func TestLookupIP_NoSuitableAddress(t *testing.T) {
+	// The net package returns this AddrError when the host resolves but has
+	// no addresses of the requested family (e.g. AAAA query for a v4-only
+	// hosts file entry). The domain exists, so this is NODATA, not SERVFAIL.
+	r := &mockResolver{results: map[string]mockLookup{
+		"ip6": {err: &net.AddrError{Err: "no suitable address found", Addr: "example.com."}},
+	}}
+
+	result := LookupIP(context.Background(), r, "ip6", "example.com.", dns.TypeAAAA)
+
+	assert.Equal(t, dns.RcodeSuccess, result.Rcode, "no suitable address should map to NODATA")
+	assert.Empty(t, result.IPs, "NODATA response should carry no addresses")
+}
+
+// TestErrNoSuitableAddressMatchesNetPackage pins our copy of the error string
+// to what the net package actually emits. A literal IP of the wrong family
+// takes the same filterAddrList path as a resolved hostname, without network
+// access.
+func TestErrNoSuitableAddressMatchesNetPackage(t *testing.T) {
+	_, err := (&net.Resolver{}).LookupNetIP(context.Background(), "ip6", "192.0.2.1")
+	require.Error(t, err)
+
+	var addrErr *net.AddrError
+	require.ErrorAs(t, err, &addrErr, "wrong-family lookup should return AddrError")
+	assert.Equal(t, errNoSuitableAddress, addrErr.Err, "net package error string should match our constant")
+}
+
+func TestLookupIP_OtherAddrError(t *testing.T) {
+	r := &mockResolver{results: map[string]mockLookup{
+		"ip4": {err: &net.AddrError{Err: "some other address problem", Addr: "example.com."}},
+	}}
+
+	result := LookupIP(context.Background(), r, "ip4", "example.com.", dns.TypeA)
+
+	assert.Equal(t, dns.RcodeServerFailure, result.Rcode, "unrecognized AddrError should map to SERVFAIL")
+}
+
+func TestLookupIP_NotFoundNXDomain(t *testing.T) {
+	r := &mockResolver{results: map[string]mockLookup{
+		"ip4": {err: &net.DNSError{Err: "no such host", Name: "example.com.", IsNotFound: true}},
+		"ip6": {err: &net.DNSError{Err: "no such host", Name: "example.com.", IsNotFound: true}},
+	}}
+
+	result := LookupIP(context.Background(), r, "ip4", "example.com.", dns.TypeA)
+
+	assert.Equal(t, dns.RcodeNameError, result.Rcode, "not found for both families should map to NXDOMAIN")
+}
+
+func TestLookupIP_NotFoundNoData(t *testing.T) {
+	r := &mockResolver{results: map[string]mockLookup{
+		"ip6": {err: &net.DNSError{Err: "no such host", Name: "example.com.", IsNotFound: true}},
+		"ip4": {ips: []netip.Addr{netip.MustParseAddr("192.0.2.1")}},
+	}}
+
+	result := LookupIP(context.Background(), r, "ip6", "example.com.", dns.TypeAAAA)
+
+	assert.Equal(t, dns.RcodeSuccess, result.Rcode, "not found with the other family present should map to NODATA")
+}
+
+func TestLookupIP_GenericError(t *testing.T) {
+	r := &mockResolver{results: map[string]mockLookup{
+		"ip4": {err: errors.New("connection refused")},
+	}}
+
+	result := LookupIP(context.Background(), r, "ip4", "example.com.", dns.TypeA)
+
+	assert.Equal(t, dns.RcodeServerFailure, result.Rcode, "generic error should map to SERVFAIL")
+}
+
+func TestLookupIP_DNSErrorNotIsNotFound(t *testing.T) {
+	r := &mockResolver{results: map[string]mockLookup{
+		"ip4": {err: &net.DNSError{Err: "server misbehaving", Name: "example.com.", IsTemporary: true}},
+	}}
+
+	result := LookupIP(context.Background(), r, "ip4", "example.com.", dns.TypeA)
+
+	assert.Equal(t, dns.RcodeServerFailure, result.Rcode, "upstream failure should map to SERVFAIL")
+}
+
+func TestStripOPT(t *testing.T) {
+	rm := &dns.Msg{
+		Extra: []dns.RR{
+			&dns.OPT{Hdr: dns.RR_Header{Name: ".", Rrtype: dns.TypeOPT}},
+			&dns.A{Hdr: dns.RR_Header{Name: "x.", Rrtype: dns.TypeA}, A: net.IPv4(1, 2, 3, 4)},
+		},
+	}
+	StripOPT(rm)
+	assert.Len(t, rm.Extra, 1, "OPT should be removed, A kept")
+	_, isOPT := rm.Extra[0].(*dns.OPT)
+	assert.False(t, isOPT, "remaining record must not be OPT")
+}
+
+func TestExtractEDE(t *testing.T) {
+	t.Run("no edns", func(t *testing.T) {
+		_, ok := ExtractEDE(&dns.Msg{})
+		assert.False(t, ok, "message without OPT has no EDE")
+	})
+
+	t.Run("edns without ede", func(t *testing.T) {
+		rm := &dns.Msg{}
+		rm.SetEdns0(4096, false)
+		_, ok := ExtractEDE(rm)
+		assert.False(t, ok, "OPT without EDE option returns false")
+	})
+
+	t.Run("with ede", func(t *testing.T) {
+		rm := &dns.Msg{}
+		opt := &dns.OPT{Hdr: dns.RR_Header{Name: ".", Rrtype: dns.TypeOPT}}
+		opt.Option = append(opt.Option, &dns.EDNS0_EDE{InfoCode: 49152, ExtraText: "upstream timeout"})
+		rm.Extra = append(rm.Extra, opt)
+
+		ede, ok := ExtractEDE(rm)
+		assert.True(t, ok, "EDE option should be found")
+		assert.Equal(t, uint16(49152), ede.InfoCode)
+		assert.Equal(t, "upstream timeout", ede.ExtraText)
+	})
+}

client/internal/dns/server.go

@@ -6,6 +6,7 @@ import (
 	"fmt"
 	"net/netip"
 	"net/url"
+	"os"
 	"slices"
 	"strings"
 	"sync"
@@ -38,11 +39,15 @@ const (
 	// defaultWarningDelayBase is the starting grace window before a
 	// "Nameserver group unreachable" event fires for a group that's
 	// never been healthy and only has overlay upstreams with no
-	// Connected peer. Per-server and overridable; see warningDelayFor.
+	// Connected peer. Per-server and overridable via envWarningDelay;
-	defaultWarningDelayBase = 30 * time.Second
+	// see warningDelay.
+	defaultWarningDelayBase = 60 * time.Second
 	// warningDelayBonusCap caps the route-count bonus added to the
-	// base grace window. See warningDelayFor.
+	// base grace window. See warningDelay.
 	warningDelayBonusCap = 30 * time.Second
+	// envWarningDelay overrides defaultWarningDelayBase with a Go duration
+	// string (e.g. "90s", "2m"). Invalid or non-positive values are ignored.
+	envWarningDelay = "NB_DNS_HEALTH_WARNING_DELAY"
 )
 
 // errNoUsableNameservers signals that a merged-domain group has no usable
@@ -135,7 +140,7 @@ type DefaultServer struct {
 	disableSys         bool
 	mux                sync.Mutex
 	service            service
-	dnsMuxMap          registeredHandlerMap
+	dnsMuxHandlers     []handlerWrapper
 	localResolver      *local.Resolver
 	wgInterface        WGIface
 	hostManager        hostManager
@@ -199,8 +204,6 @@ type handlerWrapper struct {
 	priority int
 }
 
-type registeredHandlerMap map[types.HandlerID]handlerWrapper
-
 // DefaultServerConfig holds configuration parameters for NewDefaultServer
 type DefaultServerConfig struct {
 	WgInterface    WGIface
@@ -289,7 +292,6 @@ func newDefaultServer(
 		service:           dnsService,
 		handlerChain:      handlerChain,
 		extraDomains:      make(map[domain.Domain]int),
-		dnsMuxMap:         make(registeredHandlerMap),
 		localResolver:     local.NewResolver(),
 		wgInterface:       wgInterface,
 		statusRecorder:    statusRecorder,
@@ -298,7 +300,7 @@ func newDefaultServer(
 		hostManager:       &noopHostConfigurator{},
 		mgmtCacheResolver: mgmtCacheResolver,
 		currentConfigHash: ^uint64(0), // Initialize to max uint64 to ensure first config is always applied
-		warningDelayBase:  defaultWarningDelayBase,
+		warningDelayBase:  warningDelayBaseFromEnv(),
 		healthRefresh:     make(chan struct{}, 1),
 	}
 	// Wire the local resolver against the peer status recorder so it can
@@ -328,7 +330,7 @@ func (s *DefaultServer) SetRouteSources(selected, active func() route.HAMap) {
 	type routeSettable interface {
 		setSelectedRoutes(func() route.HAMap)
 	}
-	for _, entry := range s.dnsMuxMap {
+	for _, entry := range s.dnsMuxHandlers {
 		if h, ok := entry.handler.(routeSettable); ok {
 			h.setSelectedRoutes(selected)
 		}
@@ -777,13 +779,24 @@ func (s *DefaultServer) applyHostConfig() {
 // context is released rather than leaked until GC.
 func (s *DefaultServer) registerFallback() {
 	originalNameservers := s.hostManager.getOriginalNameservers()
-	if len(originalNameservers) == 0 {
+
+	serverIP := s.service.RuntimeIP()
+	var servers []netip.AddrPort
+	for _, ns := range originalNameservers {
+		if ns == serverIP {
+			log.Debugf("skipping original nameserver %s as it is the same as the server IP %s", ns, serverIP)
+			continue
+		}
+		servers = append(servers, netip.AddrPortFrom(ns, DefaultPort))
+	}
+
+	if len(servers) == 0 {
 		log.Debugf("no fallback upstreams to register; clearing PriorityFallback handler")
 		s.clearFallback()
 		return
 	}
 
-	log.Infof("registering original nameservers %v as upstream handlers with priority %d", originalNameservers, PriorityFallback)
+	log.Infof("registering original nameservers %v as upstream handlers with priority %d", servers, PriorityFallback)
 
 	handler, err := newUpstreamResolver(
 		s.ctx,
@@ -797,11 +810,6 @@ func (s *DefaultServer) registerFallback() {
 		return
 	}
 	handler.selectedRoutes = s.selectedRoutes
-
-	var servers []netip.AddrPort
-	for _, ns := range originalNameservers {
-		servers = append(servers, netip.AddrPortFrom(ns, DefaultPort))
-	}
 	handler.addRace(servers)
 
 	prev := s.fallbackHandler
@@ -972,19 +980,23 @@ func (s *DefaultServer) usableNameServers(nameServers []nbdns.NameServer) []neti
 
 func (s *DefaultServer) updateMux(muxUpdates []handlerWrapper) {
 	// this will introduce a short period of time when the server is not able to handle DNS requests
-	for _, existing := range s.dnsMuxMap {
+	for _, existing := range s.dnsMuxHandlers {
 		s.deregisterHandler([]string{existing.domain}, existing.priority)
-		existing.handler.Stop()
+		// The local resolver is a persistent singleton shared by every custom
+		// zone and reused across config updates. Its chain registrations are
+		// per-config and must be deregistered, but Stop() cancels its lookup
+		// context (breaking external CNAME-target resolution) and clears its
+		// records, so it must not be torn down here.
+		if existing.handler != s.localResolver {
+			existing.handler.Stop()
+		}
 	}
 
-	muxUpdateMap := make(registeredHandlerMap)
-
 	for _, update := range muxUpdates {
 		s.registerHandler([]string{update.domain}, update.handler, update.priority)
-		muxUpdateMap[update.handler.ID()] = update
 	}
 
-	s.dnsMuxMap = muxUpdateMap
+	s.dnsMuxHandlers = muxUpdates
 }
 
 // updateNSGroupStates records the new group set and pokes the refresher.
@@ -1148,6 +1160,26 @@ func (s *DefaultServer) projectUnhealthy(p *nsGroupProj, servers []netip.AddrPor
 	return false
 }
 
+// warningDelayBaseFromEnv returns the base grace window, honoring
+// envWarningDelay when it holds a valid positive Go duration. Invalid or
+// non-positive values fall back to defaultWarningDelayBase.
+func warningDelayBaseFromEnv() time.Duration {
+	val := os.Getenv(envWarningDelay)
+	if val == "" {
+		return defaultWarningDelayBase
+	}
+	d, err := time.ParseDuration(val)
+	if err != nil {
+		log.Warnf("invalid %s value %q, using default %v: %v", envWarningDelay, val, defaultWarningDelayBase, err)
+		return defaultWarningDelayBase
+	}
+	if d <= 0 {
+		log.Warnf("%s must be positive, got %v, using default %v", envWarningDelay, d, defaultWarningDelayBase)
+		return defaultWarningDelayBase
+	}
+	return d
+}
+
 // warningDelay returns the grace window for the given selected-route
 // count. Scales gently: +1s per 100 routes, capped by
 // warningDelayBonusCap. Parallel handshakes mean handshake time grows
@@ -1198,7 +1230,7 @@ func (s *DefaultServer) groupHasImmediateUpstream(servers []netip.AddrPort, snap
 // in more than one handler.
 func (s *DefaultServer) collectUpstreamHealth() map[netip.AddrPort]UpstreamHealth {
 	merged := make(map[netip.AddrPort]UpstreamHealth)
-	for _, entry := range s.dnsMuxMap {
+	for _, entry := range s.dnsMuxHandlers {
 		reporter, ok := entry.handler.(upstreamHealthReporter)
 		if !ok {
 			continue

client/internal/dns/server_privileged_test.go

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

client/internal/dns/server_test.go

@@ -10,7 +10,6 @@ import (
 	"testing"
 	"time"
 
-	"github.com/golang/mock/gomock"
 	"github.com/miekg/dns"
 	log "github.com/sirupsen/logrus"
 	"github.com/stretchr/testify/assert"
@@ -23,7 +22,6 @@ import (
 	"github.com/netbirdio/netbird/client/iface"
 	"github.com/netbirdio/netbird/client/iface/configurer"
 	"github.com/netbirdio/netbird/client/iface/device"
-	pfmock "github.com/netbirdio/netbird/client/iface/mocks"
 	"github.com/netbirdio/netbird/client/iface/wgaddr"
 	"github.com/netbirdio/netbird/client/internal/dns/local"
 	"github.com/netbirdio/netbird/client/internal/dns/test"
@@ -104,481 +102,6 @@ func init() {
 	formatter.SetTextFormatter(log.StandardLogger())
 }
 
-func generateDummyHandler(d string, servers []nbdns.NameServer) *upstreamResolverBase {
-	var srvs []netip.AddrPort
-	for _, srv := range servers {
-		srvs = append(srvs, srv.AddrPort())
-	}
-	u := &upstreamResolverBase{
-		domain: domain.Domain(d),
-		cancel: func() {},
-	}
-	u.addRace(srvs)
-	return u
-}
-
-func TestUpdateDNSServer(t *testing.T) {
-
-	nameServers := []nbdns.NameServer{
-		{
-			IP:     netip.MustParseAddr("8.8.8.8"),
-			NSType: nbdns.UDPNameServerType,
-			Port:   53,
-		},
-		{
-			IP:     netip.MustParseAddr("8.8.4.4"),
-			NSType: nbdns.UDPNameServerType,
-			Port:   53,
-		},
-	}
-
-	dummyHandler := local.NewResolver()
-
-	testCases := []struct {
-		name                string
-		initUpstreamMap     registeredHandlerMap
-		initLocalZones      []nbdns.CustomZone
-		initSerial          uint64
-		inputSerial         uint64
-		inputUpdate         nbdns.Config
-		shouldFail          bool
-		expectedUpstreamMap registeredHandlerMap
-		expectedLocalQs     []dns.Question
-	}{
-		{
-			name:            "Initial Config Should Succeed",
-			initUpstreamMap: make(registeredHandlerMap),
-			initSerial:      0,
-			inputSerial:     1,
-			inputUpdate: nbdns.Config{
-				ServiceEnable: true,
-				CustomZones: []nbdns.CustomZone{
-					{
-						Domain:  "netbird.cloud",
-						Records: zoneRecords,
-					},
-				},
-				NameServerGroups: []*nbdns.NameServerGroup{
-					{
-						Domains:     []string{"netbird.io"},
-						NameServers: nameServers,
-					},
-					{
-						NameServers: nameServers,
-						Primary:     true,
-					},
-				},
-			},
-			expectedUpstreamMap: registeredHandlerMap{
-				generateDummyHandler("netbird.io", nameServers).ID(): handlerWrapper{
-					domain:   "netbird.io",
-					handler:  dummyHandler,
-					priority: PriorityUpstream,
-				},
-				dummyHandler.ID(): handlerWrapper{
-					domain:   "netbird.cloud",
-					handler:  dummyHandler,
-					priority: PriorityLocal,
-				},
-				generateDummyHandler(".", nameServers).ID(): handlerWrapper{
-					domain:   nbdns.RootZone,
-					handler:  dummyHandler,
-					priority: PriorityDefault,
-				},
-			},
-			expectedLocalQs: []dns.Question{{Name: "peera.netbird.cloud.", Qtype: dns.TypeA, Qclass: dns.ClassINET}},
-		},
-		{
-			name:           "New Config Should Succeed",
-			initLocalZones: []nbdns.CustomZone{{Domain: "netbird.cloud", Records: []nbdns.SimpleRecord{{Name: "netbird.cloud", Type: 1, Class: nbdns.DefaultClass, TTL: 300, RData: "10.0.0.1"}}}},
-			initUpstreamMap: registeredHandlerMap{
-				generateDummyHandler(zoneRecords[0].Name, nameServers).ID(): handlerWrapper{
-					domain:   "netbird.cloud",
-					handler:  dummyHandler,
-					priority: PriorityUpstream,
-				},
-			},
-			initSerial:  0,
-			inputSerial: 1,
-			inputUpdate: nbdns.Config{
-				ServiceEnable: true,
-				CustomZones: []nbdns.CustomZone{
-					{
-						Domain:  "netbird.cloud",
-						Records: zoneRecords,
-					},
-				},
-				NameServerGroups: []*nbdns.NameServerGroup{
-					{
-						Domains:     []string{"netbird.io"},
-						NameServers: nameServers,
-					},
-				},
-			},
-			expectedUpstreamMap: registeredHandlerMap{
-				generateDummyHandler("netbird.io", nameServers).ID(): handlerWrapper{
-					domain:   "netbird.io",
-					handler:  dummyHandler,
-					priority: PriorityUpstream,
-				},
-				"local-resolver": handlerWrapper{
-					domain:   "netbird.cloud",
-					handler:  dummyHandler,
-					priority: PriorityLocal,
-				},
-			},
-			expectedLocalQs: []dns.Question{{Name: zoneRecords[0].Name, Qtype: 1, Qclass: 1}},
-		},
-		{
-			name:            "Smaller Config Serial Should Be Skipped",
-			initLocalZones:  []nbdns.CustomZone{},
-			initUpstreamMap: make(registeredHandlerMap),
-			initSerial:      2,
-			inputSerial:     1,
-			shouldFail:      true,
-		},
-		{
-			name:            "Empty NS Group Domain Or Not Primary Element Should Fail",
-			initLocalZones:  []nbdns.CustomZone{},
-			initUpstreamMap: make(registeredHandlerMap),
-			initSerial:      0,
-			inputSerial:     1,
-			inputUpdate: nbdns.Config{
-				ServiceEnable: true,
-				CustomZones: []nbdns.CustomZone{
-					{
-						Domain:  "netbird.cloud",
-						Records: zoneRecords,
-					},
-				},
-				NameServerGroups: []*nbdns.NameServerGroup{
-					{
-						NameServers: nameServers,
-					},
-				},
-			},
-			shouldFail: true,
-		},
-		{
-			name:            "Invalid NS Group Nameservers list Should Fail",
-			initLocalZones:  []nbdns.CustomZone{},
-			initUpstreamMap: make(registeredHandlerMap),
-			initSerial:      0,
-			inputSerial:     1,
-			inputUpdate: nbdns.Config{
-				ServiceEnable: true,
-				CustomZones: []nbdns.CustomZone{
-					{
-						Domain:  "netbird.cloud",
-						Records: zoneRecords,
-					},
-				},
-				NameServerGroups: []*nbdns.NameServerGroup{
-					{
-						NameServers: nameServers,
-					},
-				},
-			},
-			shouldFail: true,
-		},
-		{
-			name:            "Invalid Custom Zone Records list Should Skip",
-			initLocalZones:  []nbdns.CustomZone{},
-			initUpstreamMap: make(registeredHandlerMap),
-			initSerial:      0,
-			inputSerial:     1,
-			inputUpdate: nbdns.Config{
-				ServiceEnable: true,
-				CustomZones: []nbdns.CustomZone{
-					{
-						Domain: "netbird.cloud",
-					},
-				},
-				NameServerGroups: []*nbdns.NameServerGroup{
-					{
-						NameServers: nameServers,
-						Primary:     true,
-					},
-				},
-			},
-			expectedUpstreamMap: registeredHandlerMap{generateDummyHandler(".", nameServers).ID(): handlerWrapper{
-				domain:   ".",
-				handler:  dummyHandler,
-				priority: PriorityDefault,
-			}},
-		},
-		{
-			name:           "Empty Config Should Succeed and Clean Maps",
-			initLocalZones: []nbdns.CustomZone{{Domain: "netbird.cloud", Records: []nbdns.SimpleRecord{{Name: "netbird.cloud", Type: int(dns.TypeA), Class: nbdns.DefaultClass, TTL: 300, RData: "10.0.0.1"}}}},
-			initUpstreamMap: registeredHandlerMap{
-				generateDummyHandler(zoneRecords[0].Name, nameServers).ID(): handlerWrapper{
-					domain:   zoneRecords[0].Name,
-					handler:  dummyHandler,
-					priority: PriorityUpstream,
-				},
-			},
-			initSerial:          0,
-			inputSerial:         1,
-			inputUpdate:         nbdns.Config{ServiceEnable: true},
-			expectedUpstreamMap: make(registeredHandlerMap),
-			expectedLocalQs:     []dns.Question{},
-		},
-		{
-			name:           "Disabled Service Should clean map",
-			initLocalZones: []nbdns.CustomZone{{Domain: "netbird.cloud", Records: []nbdns.SimpleRecord{{Name: "netbird.cloud", Type: int(dns.TypeA), Class: nbdns.DefaultClass, TTL: 300, RData: "10.0.0.1"}}}},
-			initUpstreamMap: registeredHandlerMap{
-				generateDummyHandler(zoneRecords[0].Name, nameServers).ID(): handlerWrapper{
-					domain:   zoneRecords[0].Name,
-					handler:  dummyHandler,
-					priority: PriorityUpstream,
-				},
-			},
-			initSerial:          0,
-			inputSerial:         1,
-			inputUpdate:         nbdns.Config{ServiceEnable: false},
-			expectedUpstreamMap: make(registeredHandlerMap),
-			expectedLocalQs:     []dns.Question{},
-		},
-	}
-
-	for n, testCase := range testCases {
-		t.Run(testCase.name, func(t *testing.T) {
-			privKey, _ := wgtypes.GenerateKey()
-			newNet, err := stdnet.NewNet(context.Background(), nil)
-			if err != nil {
-				t.Fatal(err)
-			}
-
-			opts := iface.WGIFaceOpts{
-				IFaceName:    fmt.Sprintf("utun230%d", n),
-				Address:      wgaddr.MustParseWGAddress(fmt.Sprintf("100.66.100.%d/32", n+1)),
-				WGPort:       33100,
-				WGPrivKey:    privKey.String(),
-				MTU:          iface.DefaultMTU,
-				TransportNet: newNet,
-			}
-
-			wgIface, err := iface.NewWGIFace(opts)
-			if err != nil {
-				t.Fatal(err)
-			}
-			err = wgIface.Create()
-			if err != nil {
-				t.Fatal(err)
-			}
-			defer func() {
-				err = wgIface.Close()
-				if err != nil {
-					t.Log(err)
-				}
-			}()
-			dnsServer, err := NewDefaultServer(context.Background(), DefaultServerConfig{
-				WgInterface:    wgIface,
-				CustomAddress:  "",
-				StatusRecorder: peer.NewRecorder("mgm"),
-				StateManager:   nil,
-				DisableSys:     false,
-			})
-			if err != nil {
-				t.Fatal(err)
-			}
-			err = dnsServer.Initialize()
-			if err != nil {
-				t.Fatal(err)
-			}
-			defer func() {
-				err = dnsServer.hostManager.restoreHostDNS()
-				if err != nil {
-					t.Log(err)
-				}
-			}()
-
-			dnsServer.dnsMuxMap = testCase.initUpstreamMap
-			dnsServer.localResolver.Update(testCase.initLocalZones)
-			dnsServer.updateSerial = testCase.initSerial
-
-			err = dnsServer.UpdateDNSServer(testCase.inputSerial, testCase.inputUpdate)
-			if err != nil {
-				if testCase.shouldFail {
-					return
-				}
-				t.Fatalf("update dns server should not fail, got error: %v", err)
-			}
-
-			if len(dnsServer.dnsMuxMap) != len(testCase.expectedUpstreamMap) {
-				t.Fatalf("update upstream failed, map size is different than expected, want %d, got %d", len(testCase.expectedUpstreamMap), len(dnsServer.dnsMuxMap))
-			}
-
-			for key := range testCase.expectedUpstreamMap {
-				_, found := dnsServer.dnsMuxMap[key]
-				if !found {
-					t.Fatalf("update upstream failed, key %s was not found in the dnsMuxMap: %#v", key, dnsServer.dnsMuxMap)
-				}
-			}
-
-			var responseMSG *dns.Msg
-			responseWriter := &test.MockResponseWriter{
-				WriteMsgFunc: func(m *dns.Msg) error {
-					responseMSG = m
-					return nil
-				},
-			}
-			for _, q := range testCase.expectedLocalQs {
-				dnsServer.localResolver.ServeDNS(responseWriter, &dns.Msg{
-					Question: []dns.Question{q},
-				})
-			}
-
-			if len(testCase.expectedLocalQs) > 0 {
-				assert.NotNil(t, responseMSG, "response message should not be nil")
-				assert.Equal(t, dns.RcodeSuccess, responseMSG.Rcode, "response code should be success")
-				assert.NotEmpty(t, responseMSG.Answer, "response message should have answers")
-			}
-		})
-	}
-}
-
-func TestDNSFakeResolverHandleUpdates(t *testing.T) {
-	ov := os.Getenv("NB_WG_KERNEL_DISABLED")
-	defer t.Setenv("NB_WG_KERNEL_DISABLED", ov)
-
-	t.Setenv("NB_WG_KERNEL_DISABLED", "true")
-	newNet, err := stdnet.NewNet(context.Background(), []string{"utun2301"})
-	if err != nil {
-		t.Errorf("create stdnet: %v", err)
-		return
-	}
-
-	privKey, _ := wgtypes.GeneratePrivateKey()
-	opts := iface.WGIFaceOpts{
-		IFaceName:    "utun2301",
-		Address:      wgaddr.MustParseWGAddress("100.66.100.1/32"),
-		WGPort:       33100,
-		WGPrivKey:    privKey.String(),
-		MTU:          iface.DefaultMTU,
-		TransportNet: newNet,
-	}
-	wgIface, err := iface.NewWGIFace(opts)
-	if err != nil {
-		t.Errorf("build interface wireguard: %v", err)
-		return
-	}
-
-	err = wgIface.Create()
-	if err != nil {
-		t.Errorf("create and init wireguard interface: %v", err)
-		return
-	}
-	defer func() {
-		if err = wgIface.Close(); err != nil {
-			t.Logf("close wireguard interface: %v", err)
-		}
-	}()
-
-	ctrl := gomock.NewController(t)
-	defer ctrl.Finish()
-
-	packetfilter := pfmock.NewMockPacketFilter(ctrl)
-	packetfilter.EXPECT().FilterOutbound(gomock.Any(), gomock.Any()).AnyTimes()
-	packetfilter.EXPECT().SetUDPPacketHook(gomock.Any(), gomock.Any(), gomock.Any()).AnyTimes()
-	packetfilter.EXPECT().SetTCPPacketHook(gomock.Any(), gomock.Any(), gomock.Any()).AnyTimes()
-
-	if err := wgIface.SetFilter(packetfilter); err != nil {
-		t.Errorf("set packet filter: %v", err)
-		return
-	}
-
-	dnsServer, err := NewDefaultServer(context.Background(), DefaultServerConfig{
-		WgInterface:    wgIface,
-		CustomAddress:  "",
-		StatusRecorder: peer.NewRecorder("mgm"),
-		StateManager:   nil,
-		DisableSys:     false,
-	})
-	if err != nil {
-		t.Errorf("create DNS server: %v", err)
-		return
-	}
-
-	err = dnsServer.Initialize()
-	if err != nil {
-		t.Errorf("run DNS server: %v", err)
-		return
-	}
-	defer func() {
-		if err = dnsServer.hostManager.restoreHostDNS(); err != nil {
-			t.Logf("restore DNS settings on the host: %v", err)
-			return
-		}
-	}()
-
-	dnsServer.dnsMuxMap = registeredHandlerMap{
-		"id1": handlerWrapper{
-			domain:   zoneRecords[0].Name,
-			handler:  &local.Resolver{},
-			priority: PriorityUpstream,
-		},
-	}
-	dnsServer.localResolver.Update([]nbdns.CustomZone{{Domain: "netbird.cloud", Records: []nbdns.SimpleRecord{{Name: "netbird.cloud", Type: int(dns.TypeA), Class: nbdns.DefaultClass, TTL: 300, RData: "10.0.0.1"}}}})
-	dnsServer.updateSerial = 0
-
-	nameServers := []nbdns.NameServer{
-		{
-			IP:     netip.MustParseAddr("8.8.8.8"),
-			NSType: nbdns.UDPNameServerType,
-			Port:   53,
-		},
-		{
-			IP:     netip.MustParseAddr("8.8.4.4"),
-			NSType: nbdns.UDPNameServerType,
-			Port:   53,
-		},
-	}
-
-	update := nbdns.Config{
-		ServiceEnable: true,
-		CustomZones: []nbdns.CustomZone{
-			{
-				Domain:  "netbird.cloud",
-				Records: zoneRecords,
-			},
-		},
-		NameServerGroups: []*nbdns.NameServerGroup{
-			{
-				Domains:     []string{"netbird.io"},
-				NameServers: nameServers,
-			},
-			{
-				NameServers: nameServers,
-				Primary:     true,
-			},
-		},
-	}
-
-	// Start the server with regular configuration
-	if err := dnsServer.UpdateDNSServer(1, update); err != nil {
-		t.Fatalf("update dns server should not fail, got error: %v", err)
-		return
-	}
-
-	update2 := update
-	update2.ServiceEnable = false
-	// Disable the server, stop the listener
-	if err := dnsServer.UpdateDNSServer(2, update2); err != nil {
-		t.Fatalf("update dns server should not fail, got error: %v", err)
-		return
-	}
-
-	update3 := update2
-	update3.NameServerGroups = update3.NameServerGroups[:1]
-	// But service still get updates and we checking that we handle
-	// internal state in the right way
-	if err := dnsServer.UpdateDNSServer(3, update3); err != nil {
-		t.Fatalf("update dns server should not fail, got error: %v", err)
-		return
-	}
-}
-
 func TestDNSServerStartStop(t *testing.T) {
 	testCases := []struct {
 		name     string
@@ -1029,15 +552,15 @@ func (m *mockService) RegisterMux(string, dns.Handler) {}
 func (m *mockService) DeregisterMux(string)            {}
 
 func TestDefaultServer_UpdateMux(t *testing.T) {
-	baseMatchHandlers := registeredHandlerMap{
+	baseMatchHandlers := []handlerWrapper{
-		"upstream-group1": {
+		{
 			domain: "example.com",
 			handler: &mockHandler{
 				Id: "upstream-group1",
 			},
 			priority: PriorityUpstream,
 		},
-		"upstream-group2": {
+		{
 			domain: "example.com",
 			handler: &mockHandler{
 				Id: "upstream-group2",
@@ -1046,15 +569,15 @@ func TestDefaultServer_UpdateMux(t *testing.T) {
 		},
 	}
 
-	baseRootHandlers := registeredHandlerMap{
+	baseRootHandlers := []handlerWrapper{
-		"upstream-root1": {
+		{
 			domain: ".",
 			handler: &mockHandler{
 				Id: "upstream-root1",
 			},
 			priority: PriorityDefault,
 		},
-		"upstream-root2": {
+		{
 			domain: ".",
 			handler: &mockHandler{
 				Id: "upstream-root2",
@@ -1063,22 +586,22 @@ func TestDefaultServer_UpdateMux(t *testing.T) {
 		},
 	}
 
-	baseMixedHandlers := registeredHandlerMap{
+	baseMixedHandlers := []handlerWrapper{
-		"upstream-group1": {
+		{
 			domain: "example.com",
 			handler: &mockHandler{
 				Id: "upstream-group1",
 			},
 			priority: PriorityUpstream,
 		},
-		"upstream-group2": {
+		{
 			domain: "example.com",
 			handler: &mockHandler{
 				Id: "upstream-group2",
 			},
 			priority: PriorityUpstream - 1,
 		},
-		"upstream-other": {
+		{
 			domain: "other.com",
 			handler: &mockHandler{
 				Id: "upstream-other",
@@ -1089,7 +612,7 @@ func TestDefaultServer_UpdateMux(t *testing.T) {
 
 	tests := []struct {
 		name             string
-		initialHandlers  registeredHandlerMap
+		initialHandlers  []handlerWrapper
 		updates          []handlerWrapper
 		expectedHandlers map[string]string // map[HandlerID]domain
 		description      string
@@ -1373,32 +896,38 @@ func TestDefaultServer_UpdateMux(t *testing.T) {
 	for _, tt := range tests {
 		t.Run(tt.name, func(t *testing.T) {
 			server := &DefaultServer{
-				dnsMuxMap:    tt.initialHandlers,
+				dnsMuxHandlers: tt.initialHandlers,
-				handlerChain: NewHandlerChain(),
+				handlerChain:   NewHandlerChain(),
-				service:      &mockService{},
+				service:        &mockService{},
 			}
 
 			// Perform the update
 			server.updateMux(tt.updates)
 
 			// Verify the results
-			assert.Equal(t, len(tt.expectedHandlers), len(server.dnsMuxMap),
+			assert.Equal(t, len(tt.expectedHandlers), len(server.dnsMuxHandlers),
 				"Number of handlers after update doesn't match expected")
 
 			// Check each expected handler
 			for id, expectedDomain := range tt.expectedHandlers {
-				handler, exists := server.dnsMuxMap[types.HandlerID(id)]
+				var found *handlerWrapper
-				assert.True(t, exists, "Expected handler %s not found", id)
+				for i := range server.dnsMuxHandlers {
-				if exists {
+					if server.dnsMuxHandlers[i].handler.ID() == types.HandlerID(id) {
-					assert.Equal(t, expectedDomain, handler.domain,
+						found = &server.dnsMuxHandlers[i]
+						break
+					}
+				}
+				assert.NotNil(t, found, "Expected handler %s not found", id)
+				if found != nil {
+					assert.Equal(t, expectedDomain, found.domain,
 						"Domain mismatch for handler %s", id)
 				}
 			}
 
 			// Verify no unexpected handlers exist
-			for HandlerID := range server.dnsMuxMap {
+			for _, entry := range server.dnsMuxHandlers {
-				_, expected := tt.expectedHandlers[string(HandlerID)]
+				_, expected := tt.expectedHandlers[string(entry.handler.ID())]
-				assert.True(t, expected, "Unexpected handler found: %s", HandlerID)
+				assert.True(t, expected, "Unexpected handler found: %s", entry.handler.ID())
 			}
 
 			// Verify the handlerChain state and order
@@ -1413,7 +942,7 @@ func TestDefaultServer_UpdateMux(t *testing.T) {
 
 				// Verify handler exists in mux
 				foundInMux := false
-				for _, muxEntry := range server.dnsMuxMap {
+				for _, muxEntry := range server.dnsMuxHandlers {
 					if chainEntry.Handler == muxEntry.handler &&
 						chainEntry.Priority == muxEntry.priority &&
 						chainEntry.Pattern == dns.Fqdn(muxEntry.domain) {
@@ -1422,12 +951,108 @@ func TestDefaultServer_UpdateMux(t *testing.T) {
 					}
 				}
 				assert.True(t, foundInMux,
-					"Handler in chain not found in dnsMuxMap")
+					"Handler in chain not found in dnsMuxHandlers")
 			}
 		})
 	}
 }
 
+// chainHasPattern reports whether the handler chain holds an entry registered
+// for the given fqdn pattern at the given priority.
+func chainHasPattern(s *DefaultServer, pattern string, priority int) bool {
+	for _, h := range s.handlerChain.handlers {
+		if h.OrigPattern == pattern && h.Priority == priority {
+			return true
+		}
+	}
+	return false
+}
+
+// TestDefaultServer_UpdateMux_SharedHandlerZoneRemoval verifies that updateMux
+// tracks each (handler, domain) registration independently when one handler
+// serves multiple zones. Every custom zone is served by the same handler
+// instance (the local resolver, whose ID is the constant "local-resolver"), so
+// removing one zone must deregister exactly that zone's chain entry and leave
+// the others in place. Tracking registrations by handler ID alone collapses all
+// zones onto one entry, leaving removed zones in the chain to answer
+// authoritatively with no records.
+func TestDefaultServer_UpdateMux_SharedHandlerZoneRemoval(t *testing.T) {
+	// One handler serves every custom zone, mirroring s.localResolver.
+	shared := &mockHandler{Id: "local-resolver"}
+
+	server := &DefaultServer{
+		handlerChain: NewHandlerChain(),
+		service:      &mockService{},
+	}
+
+	// Two custom zones under the same handler. The surviving zone is registered
+	// last, mirroring the management emission order.
+	server.updateMux([]handlerWrapper{
+		{domain: "userzone.test", handler: shared, priority: PriorityLocal},
+		{domain: "peerzone.test", handler: shared, priority: PriorityLocal},
+	})
+
+	require.True(t, chainHasPattern(server, "userzone.test.", PriorityLocal),
+		"userzone.test should be registered after the first update")
+	require.True(t, chainHasPattern(server, "peerzone.test.", PriorityLocal),
+		"peerzone.test should be registered after the first update")
+
+	// Remove one zone, keep the other.
+	server.updateMux([]handlerWrapper{
+		{domain: "peerzone.test", handler: shared, priority: PriorityLocal},
+	})
+
+	assert.True(t, chainHasPattern(server, "peerzone.test.", PriorityLocal),
+		"peerzone.test should remain after removing userzone.test")
+	assert.False(t, chainHasPattern(server, "userzone.test.", PriorityLocal),
+		"userzone.test handler must be deregistered, not leaked in the chain")
+}
+
+// TestDefaultServer_UpdateMux_PreservesLocalResolver verifies that updateMux
+// does not tear down the shared local resolver during reconfiguration. The
+// resolver is a process-lifetime singleton reused across config updates;
+// Stop() cancels its lookup context (breaking external CNAME-target
+// resolution) and clears its records. updateMux must deregister its chain
+// entries without stopping it. Records surviving a teardown update is the
+// observable proxy: Stop() would have cleared them.
+func TestDefaultServer_UpdateMux_PreservesLocalResolver(t *testing.T) {
+	resolver := local.NewResolver()
+	require.NoError(t, resolver.RegisterRecord(nbdns.SimpleRecord{
+		Name:  "peer.netbird.cloud.",
+		Type:  int(dns.TypeA),
+		Class: nbdns.DefaultClass,
+		TTL:   300,
+		RData: "10.0.0.1",
+	}))
+
+	server := &DefaultServer{
+		handlerChain:  NewHandlerChain(),
+		service:       &mockService{},
+		localResolver: resolver,
+	}
+
+	server.updateMux([]handlerWrapper{
+		{domain: "netbird.cloud", handler: resolver, priority: PriorityLocal},
+	})
+
+	// Remove the zone. The resolver must survive so its records and lookup
+	// context stay intact for the next registration.
+	server.updateMux(nil)
+
+	var response *dns.Msg
+	resolver.ServeDNS(&test.MockResponseWriter{
+		WriteMsgFunc: func(m *dns.Msg) error {
+			response = m
+			return nil
+		},
+	}, &dns.Msg{Question: []dns.Question{{Name: "peer.netbird.cloud.", Qtype: dns.TypeA, Qclass: dns.ClassINET}}})
+
+	require.NotNil(t, response, "local resolver should answer after teardown")
+	assert.Equal(t, dns.RcodeSuccess, response.Rcode,
+		"local resolver records must survive teardown; updateMux must not Stop() the shared resolver")
+	assert.NotEmpty(t, response.Answer, "answer should contain the surviving record")
+}
+
 func TestExtraDomains(t *testing.T) {
 	tests := []struct {
 		name                string
@@ -2049,7 +1674,6 @@ func TestBuildUpstreamHandler_MergesGroupsPerDomain(t *testing.T) {
 		localResolver: local.NewResolver(),
 		handlerChain:  NewHandlerChain(),
 		hostManager:   &noopHostConfigurator{},
-		dnsMuxMap:     make(registeredHandlerMap),
 	}
 
 	groups := []*nbdns.NameServerGroup{
@@ -2207,7 +1831,7 @@ func TestEvaluateNSGroupHealth(t *testing.T) {
 	}
 }
 
-// healthStubHandler is a minimal dnsMuxMap entry that exposes a fixed
+// healthStubHandler is a minimal dnsMuxHandlers entry that exposes a fixed
 // UpstreamHealth snapshot, letting tests drive recomputeNSGroupStates
 // without spinning up real handlers.
 type healthStubHandler struct {
@@ -2283,12 +1907,11 @@ func newProjTestFixture(t *testing.T) *projTestFixture {
 		ctx:              context.Background(),
 		wgInterface:      &mocWGIface{},
 		statusRecorder:   recorder,
-		dnsMuxMap:        make(registeredHandlerMap),
 		selectedRoutes:   func() route.HAMap { return fx.selected },
 		activeRoutes:     func() route.HAMap { return fx.active },
 		warningDelayBase: defaultWarningDelayBase,
 	}
-	fx.server.dnsMuxMap["example.com"] = handlerWrapper{domain: "example.com", handler: fx.stub, priority: PriorityUpstream}
+	fx.server.dnsMuxHandlers = []handlerWrapper{{domain: "example.com", handler: fx.stub, priority: PriorityUpstream}}
 
 	fx.server.mux.Lock()
 	fx.server.updateNSGroupStates([]*nbdns.NameServerGroup{fx.group})
@@ -2395,7 +2018,6 @@ func TestProjection_OverlayAddrNoRouteDelaysWarning(t *testing.T) {
 		ctx:              context.Background(),
 		wgInterface:      &mocWGIface{},
 		statusRecorder:   recorder,
-		dnsMuxMap:        make(registeredHandlerMap),
 		selectedRoutes:   func() route.HAMap { return nil },
 		activeRoutes:     func() route.HAMap { return nil },
 		warningDelayBase: 50 * time.Millisecond,
@@ -2407,7 +2029,7 @@ func TestProjection_OverlayAddrNoRouteDelaysWarning(t *testing.T) {
 	stub := &healthStubHandler{health: map[netip.AddrPort]UpstreamHealth{
 		overlayPeer: {LastFail: time.Now(), LastErr: "timeout"},
 	}}
-	server.dnsMuxMap["example.com"] = handlerWrapper{domain: "example.com", handler: stub, priority: PriorityUpstream}
+	server.dnsMuxHandlers = []handlerWrapper{{domain: "example.com", handler: stub, priority: PriorityUpstream}}
 
 	server.mux.Lock()
 	server.updateNSGroupStates([]*nbdns.NameServerGroup{group})
@@ -2444,7 +2066,6 @@ func TestProjection_StopClearsHealthState(t *testing.T) {
 		service:           NewServiceViaMemory(wgIface),
 		hostManager:       &noopHostConfigurator{},
 		extraDomains:      map[domain.Domain]int{},
-		dnsMuxMap:         make(registeredHandlerMap),
 		statusRecorder:    peer.NewRecorder("mgm"),
 		selectedRoutes:    func() route.HAMap { return nil },
 		activeRoutes:      func() route.HAMap { return nil },
@@ -2459,7 +2080,7 @@ func TestProjection_StopClearsHealthState(t *testing.T) {
 		NameServers: []nbdns.NameServer{{IP: srv.Addr(), NSType: nbdns.UDPNameServerType, Port: int(srv.Port())}},
 	}
 	stub := &healthStubHandler{health: map[netip.AddrPort]UpstreamHealth{srv: {LastOk: time.Now()}}}
-	server.dnsMuxMap["example.com"] = handlerWrapper{domain: "example.com", handler: stub, priority: PriorityUpstream}
+	server.dnsMuxHandlers = []handlerWrapper{{domain: "example.com", handler: stub, priority: PriorityUpstream}}
 
 	server.mux.Lock()
 	server.updateNSGroupStates([]*nbdns.NameServerGroup{group})
@@ -2484,6 +2105,32 @@ func TestProjection_StopClearsHealthState(t *testing.T) {
 // rule 3: startup failures while the peer is handshaking, then the peer
 // comes up and a query succeeds before the grace window elapses. No
 // warning should ever have fired, and no recovery either.
+func TestWarningDelayBaseFromEnv(t *testing.T) {
+	tests := []struct {
+		name string
+		set  bool
+		val  string
+		want time.Duration
+	}{
+		{name: "unset uses default", set: false, want: defaultWarningDelayBase},
+		{name: "valid override", set: true, val: "90s", want: 90 * time.Second},
+		{name: "valid minutes", set: true, val: "2m", want: 2 * time.Minute},
+		{name: "invalid falls back", set: true, val: "notaduration", want: defaultWarningDelayBase},
+		{name: "zero falls back", set: true, val: "0s", want: defaultWarningDelayBase},
+		{name: "negative falls back", set: true, val: "-30s", want: defaultWarningDelayBase},
+	}
+
+	for _, tc := range tests {
+		t.Run(tc.name, func(t *testing.T) {
+			t.Setenv(envWarningDelay, tc.val)
+			if !tc.set {
+				os.Unsetenv(envWarningDelay)
+			}
+			assert.Equal(t, tc.want, warningDelayBaseFromEnv(), "grace window base")
+		})
+	}
+}
+
 func TestProjection_OverlayRecoversDuringGrace(t *testing.T) {
 	fx := newProjTestFixture(t)
 	fx.server.warningDelayBase = 200 * time.Millisecond
@@ -2595,7 +2242,6 @@ func TestProjection_MixedGroupEmitsImmediately(t *testing.T) {
 	server := &DefaultServer{
 		ctx:              context.Background(),
 		statusRecorder:   recorder,
-		dnsMuxMap:        make(registeredHandlerMap),
 		selectedRoutes:   func() route.HAMap { return overlayMap },
 		activeRoutes:     func() route.HAMap { return nil },
 		warningDelayBase: time.Hour,
@@ -2613,7 +2259,7 @@ func TestProjection_MixedGroupEmitsImmediately(t *testing.T) {
 			overlay: {LastFail: time.Now(), LastErr: "timeout"},
 		},
 	}
-	server.dnsMuxMap["example.com"] = handlerWrapper{domain: "example.com", handler: stub, priority: PriorityUpstream}
+	server.dnsMuxHandlers = []handlerWrapper{{domain: "example.com", handler: stub, priority: PriorityUpstream}}
 
 	server.mux.Lock()
 	server.updateNSGroupStates([]*nbdns.NameServerGroup{group})
@@ -2640,7 +2286,6 @@ func TestDNSLoopPrevention(t *testing.T) {
 		localResolver: local.NewResolver(),
 		handlerChain:  NewHandlerChain(),
 		hostManager:   &noopHostConfigurator{},
-		dnsMuxMap:     make(registeredHandlerMap),
 	}
 
 	tests := []struct {

client/internal/dns/upstream.go

@@ -443,29 +443,32 @@ func (u *upstreamResolverBase) queryUpstream(parentCtx context.Context, r *dns.M
 		return raceResult{}, &upstreamFailure{upstream: upstream, reason: "no response"}
 	}
 
+	// A valid response means the upstream is reachable, whatever the Rcode.
+	u.markUpstreamOk(upstream)
+
 	proto := ""
 	if upstreamProto != nil {
 		proto = upstreamProto.protocol
 	}
 
 	if rm.Rcode == dns.RcodeServerFailure || rm.Rcode == dns.RcodeRefused {
+		// SERVFAIL and REFUSED are per-question outcomes (DNSSEC-bogus names,
+		// refused zones, transient recursion errors), not reachability
+		// problems: fail over for a better answer but keep the upstream healthy.
 		if code, ok := nonRetryableEDE(rm); ok {
 			if !hadEdns {
-				stripOPT(rm)
+				resutil.StripOPT(rm)
 			}
-			u.markUpstreamOk(upstream)
 			return raceResult{msg: rm, upstream: upstream, protocol: proto, ede: edeName(code)}, nil
 		}
 		reason := dns.RcodeToString[rm.Rcode]
-		u.markUpstreamFail(upstream, reason)
 		return raceResult{}, &upstreamFailure{upstream: upstream, reason: reason}
 	}
 
 	if !hadEdns {
-		stripOPT(rm)
+		resutil.StripOPT(rm)
 	}
 
-	u.markUpstreamOk(upstream)
 	return raceResult{msg: rm, upstream: upstream, protocol: proto}, nil
 }
 
@@ -520,22 +523,6 @@ func upstreamUDPSize() uint16 {
 	return dns.MinMsgSize
 }
 
-// stripOPT removes any OPT pseudo-RRs from the response's Extra section so
-// the response complies with RFC 6891 when the client did not advertise EDNS0.
-func stripOPT(rm *dns.Msg) {
-	if len(rm.Extra) == 0 {
-		return
-	}
-	out := rm.Extra[:0]
-	for _, rr := range rm.Extra {
-		if _, ok := rr.(*dns.OPT); ok {
-			continue
-		}
-		out = append(out, rr)
-	}
-	rm.Extra = out
-}
-
 func (u *upstreamResolverBase) handleUpstreamError(err error, upstream netip.AddrPort, startTime time.Time) *upstreamFailure {
 	if !errors.Is(err, context.DeadlineExceeded) && !isTimeout(err) {
 		return &upstreamFailure{upstream: upstream, reason: err.Error()}

client/internal/dns/upstream_test.go

@@ -517,6 +517,78 @@ func TestUpstreamResolver_HealthTracking(t *testing.T) {
 	assert.NotContains(t, health, bad, "sibling upstream should not be queried when primary answers")
 }
 
+// TestUpstreamResolver_HealthTracking_ResponseMeansReachable verifies that an
+// upstream which answers with SERVFAIL or REFUSED is recorded as healthy:
+// those are per-question outcomes from a reachable server and must not mark
+// the upstream unhealthy. Only transport failures (timeouts) do.
+func TestUpstreamResolver_HealthTracking_ResponseMeansReachable(t *testing.T) {
+	a := netip.MustParseAddrPort("192.0.2.10:53")
+	b := netip.MustParseAddrPort("192.0.2.11:53")
+	timeoutErr := &net.OpError{Op: "read", Err: fmt.Errorf("i/o timeout")}
+
+	tests := []struct {
+		name        string
+		respA       mockUpstreamResponse
+		respB       mockUpstreamResponse
+		wantHealthy bool
+	}{
+		{
+			name:        "both SERVFAIL are reachable",
+			respA:       mockUpstreamResponse{msg: buildMockResponse(dns.RcodeServerFailure, "")},
+			respB:       mockUpstreamResponse{msg: buildMockResponse(dns.RcodeServerFailure, "")},
+			wantHealthy: true,
+		},
+		{
+			name:        "both REFUSED are reachable",
+			respA:       mockUpstreamResponse{msg: buildMockResponse(dns.RcodeRefused, "")},
+			respB:       mockUpstreamResponse{msg: buildMockResponse(dns.RcodeRefused, "")},
+			wantHealthy: true,
+		},
+		{
+			name:        "timeout marks unhealthy",
+			respA:       mockUpstreamResponse{err: timeoutErr},
+			respB:       mockUpstreamResponse{err: timeoutErr},
+			wantHealthy: false,
+		},
+	}
+
+	for _, tc := range tests {
+		t.Run(tc.name, func(t *testing.T) {
+			mockClient := &mockUpstreamResolverPerServer{
+				responses: map[string]mockUpstreamResponse{
+					a.String(): tc.respA,
+					b.String(): tc.respB,
+				},
+				rtt: time.Millisecond,
+			}
+
+			ctx, cancel := context.WithCancel(context.Background())
+			defer cancel()
+
+			resolver := &upstreamResolverBase{
+				ctx:             ctx,
+				upstreamClient:  mockClient,
+				upstreamTimeout: UpstreamTimeout,
+			}
+			resolver.addRace([]netip.AddrPort{a, b})
+
+			responseWriter := &test.MockResponseWriter{WriteMsgFunc: func(m *dns.Msg) error { return nil }}
+			resolver.ServeDNS(responseWriter, new(dns.Msg).SetQuestion("example.com.", dns.TypeA))
+
+			health := resolver.UpstreamHealth()
+			require.Contains(t, health, a, "primary upstream should have a health record")
+			if tc.wantHealthy {
+				assert.False(t, health[a].LastOk.IsZero(), "responding upstream should have LastOk set")
+				assert.True(t, health[a].LastFail.IsZero(), "responding upstream should not be marked failed")
+				assert.Empty(t, health[a].LastErr, "responding upstream should have no error")
+			} else {
+				assert.False(t, health[a].LastFail.IsZero(), "timed-out upstream should be marked failed")
+				assert.NotEmpty(t, health[a].LastErr, "timed-out upstream should record an error")
+			}
+		})
+	}
+}
+
 func TestFormatFailures(t *testing.T) {
 	testCases := []struct {
 		name     string
@@ -913,19 +985,6 @@ func TestEDEName(t *testing.T) {
 	assert.Equal(t, "EDE 9999", edeName(9999), "unknown code falls back to numeric")
 }
 
-func TestStripOPT(t *testing.T) {
-	rm := &dns.Msg{
-		Extra: []dns.RR{
-			&dns.OPT{Hdr: dns.RR_Header{Name: ".", Rrtype: dns.TypeOPT}},
-			&dns.A{Hdr: dns.RR_Header{Name: "x.", Rrtype: dns.TypeA}, A: net.IPv4(1, 2, 3, 4)},
-		},
-	}
-	stripOPT(rm)
-	assert.Len(t, rm.Extra, 1, "OPT should be removed, A kept")
-	_, isOPT := rm.Extra[0].(*dns.OPT)
-	assert.False(t, isOPT, "remaining record must not be OPT")
-}
-
 func TestUpstreamResolver_NonRetryableEDEShortCircuits(t *testing.T) {
 	upstream1 := netip.MustParseAddrPort("192.0.2.1:53")
 	upstream2 := netip.MustParseAddrPort("192.0.2.2:53")

client/internal/dnsfwd/forwarder.go

@@ -26,6 +26,15 @@ import (
 const errResolveFailed = "failed to resolve query for domain=%s: %v"
 const upstreamTimeout = 15 * time.Second
 
+// EDE info codes the forwarder emits on upstream failures so the querying
+// client can see the reason without inspecting this peer's logs. They live in
+// the RFC 8914 Private Use range (49152-65535); the Go resolver never exposes a
+// real upstream EDE here, so these cannot collide with a genuine code.
+const (
+	edeNetbirdUpstreamTimeout uint16 = 49152
+	edeNetbirdUpstreamFailure uint16 = 49153
+)
+
 type resolver interface {
 	LookupNetIP(ctx context.Context, network, host string) ([]netip.Addr, error)
 }
@@ -220,7 +229,7 @@ func (f *DNSForwarder) handleDNSQuery(logger *log.Entry, w dns.ResponseWriter, q
 
 	result := resutil.LookupIP(ctx, f.resolver, network, qname, question.Qtype)
 	if result.Err != nil {
-		f.handleDNSError(ctx, logger, w, question, resp, qname, result, startTime)
+		f.handleDNSError(ctx, logger, w, question, resp, qname, result, query.IsEdns0() != nil, startTime)
 		return
 	}
 
@@ -333,6 +342,7 @@ func (f *DNSForwarder) handleDNSError(
 	resp *dns.Msg,
 	domain string,
 	result resutil.LookupResult,
+	reqHasEdns bool,
 	startTime time.Time,
 ) {
 	qType := question.Qtype
@@ -374,6 +384,10 @@ func (f *DNSForwarder) handleDNSError(
 		logger.Warnf(errResolveFailed, domain, result.Err)
 	}
 
+	if reqHasEdns {
+		attachEDE(resp, edeCodeFor(dnsErr), edeText(dnsErr))
+	}
+
 	f.writeResponse(logger, w, resp, domain, startTime)
 }
 
@@ -414,3 +428,33 @@ func (f *DNSForwarder) getMatchingEntries(domain string) (route.ResID, []*Forwar
 
 	return selectedResId, matches
 }
+
+// edeCodeFor maps an upstream lookup error to the NetBird EDE info code.
+func edeCodeFor(dnsErr *net.DNSError) uint16 {
+	if dnsErr != nil && dnsErr.IsTimeout {
+		return edeNetbirdUpstreamTimeout
+	}
+	return edeNetbirdUpstreamFailure
+}
+
+// edeText builds the EDE extra-text describing the class of upstream failure.
+// It deliberately omits the upstream server address, which may be an internal
+// resolver and is exposed to any client permitted to use the route; the full
+// detail stays in the forwarder's local log.
+func edeText(dnsErr *net.DNSError) string {
+	if dnsErr != nil && dnsErr.IsTimeout {
+		return "netbird forwarder: upstream timeout"
+	}
+	return "netbird forwarder: upstream failure"
+}
+
+// attachEDE adds an Extended DNS Error (RFC 8914) option to the response,
+// creating the OPT pseudo-record if the response does not already carry one.
+func attachEDE(resp *dns.Msg, code uint16, text string) {
+	opt := resp.IsEdns0()
+	if opt == nil {
+		resp.SetEdns0(dns.DefaultMsgSize, false)
+		opt = resp.IsEdns0()
+	}
+	opt.Option = append(opt.Option, &dns.EDNS0_EDE{InfoCode: code, ExtraText: text})
+}

client/internal/dnsfwd/forwarder_test.go

@@ -16,6 +16,7 @@ import (
 	"github.com/stretchr/testify/require"
 
 	firewall "github.com/netbirdio/netbird/client/firewall/manager"
+	"github.com/netbirdio/netbird/client/internal/dns/resutil"
 	"github.com/netbirdio/netbird/client/internal/dns/test"
 	"github.com/netbirdio/netbird/client/internal/peer"
 	"github.com/netbirdio/netbird/route"
@@ -617,6 +618,85 @@ func TestDNSForwarder_ResponseCodes(t *testing.T) {
 	}
 }
 
+func TestDNSForwarder_UpstreamFailureEDE(t *testing.T) {
+	tests := []struct {
+		name        string
+		lookupErr   error
+		reqEdns     bool
+		wantEDE     bool
+		wantCode    uint16
+		wantTextHas string
+	}{
+		{
+			name:        "timeout with edns0",
+			lookupErr:   &net.DNSError{Err: "i/o timeout", Server: "10.0.0.53:53", IsTimeout: true},
+			reqEdns:     true,
+			wantEDE:     true,
+			wantCode:    edeNetbirdUpstreamTimeout,
+			wantTextHas: "netbird forwarder: upstream timeout",
+		},
+		{
+			name:        "server failure with edns0",
+			lookupErr:   &net.DNSError{Err: "server misbehaving", Server: "10.0.0.53:53"},
+			reqEdns:     true,
+			wantEDE:     true,
+			wantCode:    edeNetbirdUpstreamFailure,
+			wantTextHas: "netbird forwarder: upstream failure",
+		},
+		{
+			name:      "no edns0 in request omits ede",
+			lookupErr: &net.DNSError{Err: "server misbehaving", Server: "10.0.0.53:53"},
+			reqEdns:   false,
+			wantEDE:   false,
+		},
+	}
+
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			mockResolver := &MockResolver{}
+			forwarder := NewDNSForwarder(netip.MustParseAddrPort("127.0.0.1:0"), 300, nil, &peer.Status{}, nil)
+			forwarder.resolver = mockResolver
+
+			d, err := domain.FromString("example.com")
+			require.NoError(t, err)
+			forwarder.UpdateDomains([]*ForwarderEntry{{Domain: d, ResID: "test-res"}})
+
+			mockResolver.On("LookupNetIP", mock.Anything, "ip4", "example.com.").
+				Return([]netip.Addr(nil), tt.lookupErr).Once()
+
+			query := &dns.Msg{}
+			query.SetQuestion("example.com.", dns.TypeA)
+			if tt.reqEdns {
+				query.SetEdns0(dns.DefaultMsgSize, false)
+			}
+
+			var writtenResp *dns.Msg
+			mockWriter := &test.MockResponseWriter{
+				WriteMsgFunc: func(m *dns.Msg) error {
+					writtenResp = m
+					return nil
+				},
+			}
+
+			forwarder.handleDNSQuery(log.NewEntry(log.StandardLogger()), mockWriter, query, time.Now())
+			mockResolver.AssertExpectations(t)
+
+			require.NotNil(t, writtenResp, "expected a response")
+			assert.Equal(t, dns.RcodeServerFailure, writtenResp.Rcode, "upstream failure must be SERVFAIL")
+
+			ede, ok := resutil.ExtractEDE(writtenResp)
+			if !tt.wantEDE {
+				assert.False(t, ok, "response must not carry EDE")
+				return
+			}
+			require.True(t, ok, "response must carry EDE")
+			assert.Equal(t, tt.wantCode, ede.InfoCode, "EDE info code")
+			assert.Contains(t, ede.ExtraText, tt.wantTextHas, "EDE extra-text")
+			assert.NotContains(t, ede.ExtraText, "10.0.0.53", "must not leak upstream server address")
+		})
+	}
+}
+
 func TestDNSForwarder_TCPTruncation(t *testing.T) {
 	// Test that large UDP responses are truncated with TC bit set
 	mockResolver := &MockResolver{}

client/internal/engine.go

@@ -53,7 +53,6 @@ import (
 	"github.com/netbirdio/netbird/client/internal/relay"
 	"github.com/netbirdio/netbird/client/internal/rosenpass"
 	"github.com/netbirdio/netbird/client/internal/routemanager"
-	"github.com/netbirdio/netbird/client/internal/routemanager/systemops"
 	"github.com/netbirdio/netbird/client/internal/statemanager"
 	"github.com/netbirdio/netbird/client/internal/syncstore"
 	"github.com/netbirdio/netbird/client/internal/updater"
@@ -87,6 +86,8 @@ const (
 
 var ErrResetConnection = fmt.Errorf("reset connection")
 
+var ErrEngineAlreadyStarted = errors.New("engine already started")
+
 type EngineConfig struct {
 	WgPort      int
 	WgIfaceName string
@@ -200,6 +201,8 @@ type Engine struct {
 	ctx    context.Context
 	cancel context.CancelFunc
 
+	started bool
+
 	wgInterface WGIface
 
 	udpMux *udpmux.UniversalUDPMuxDefault
@@ -207,6 +210,12 @@ type Engine struct {
 	// networkSerial is the latest CurrentSerial (state ID) of the network sent by the Management service
 	networkSerial uint64
 
+	// forwardingRules holds the ingress forward rules applied for the current target.
+	// Wholesale sections (incl. forward rules) run only on the first pass of a target;
+	// it is stashed here so the final, peer-converged pass can build the lazy-connection
+	// exclude list without recomputing them on every bounded peer pass.
+	forwardingRules []firewallManager.ForwardRule
+
 	networkMonitor *networkmonitor.NetworkMonitor
 
 	sshServer sshServer
@@ -240,7 +249,7 @@ type Engine struct {
 	syncStore    syncstore.Store
 	syncStoreDir string
 
-	flowManager         nftypes.FlowManager
+	flowManager nftypes.FlowManager
 
 	// auto-update
 	updateManager *updater.Manager
@@ -280,9 +289,15 @@ func NewEngine(
 	services EngineServices,
 	mobileDep MobileDependency,
 ) *Engine {
+	// The engine is single-use: a fresh instance is built per connection
+	// cycle (see Client.run), so the run context is created once here rather
+	// than in Start.
+	ctx, cancel := context.WithCancel(clientCtx)
 	engine := &Engine{
 		clientCtx:          clientCtx,
 		clientCancel:       clientCancel,
+		ctx:                ctx,
+		cancel:             cancel,
 		signal:             services.SignalClient,
 		signaler:           peer.NewSignaler(services.SignalClient, config.WgPrivateKey),
 		mgmClient:          services.MgmClient,
@@ -315,8 +330,34 @@ func (e *Engine) Stop() error {
 		log.Debugf("tried stopping engine that is nil")
 		return nil
 	}
+	e.cancel()
 	e.syncMsgMux.Lock()
 
+	e.stopLocked()
+
+	e.syncMsgMux.Unlock()
+
+	timeout := e.calculateShutdownTimeout()
+	log.Debugf("waiting for goroutines to finish with timeout: %v", timeout)
+	shutdownCtx, cancel := context.WithTimeout(context.Background(), timeout)
+	defer cancel()
+
+	if err := waitWithContext(shutdownCtx, &e.shutdownWg); err != nil {
+		log.Warnf("shutdown timeout exceeded after %v, some goroutines may still be running", timeout)
+	}
+
+	log.Infof("stopped Netbird Engine")
+
+	return nil
+}
+
+// stopLocked tears down everything Start may have brought up, in the order
+// teardown requires (DNS before the interface goes down, flow manager after).
+// The caller must hold syncMsgMux. It is shared by Stop and by Start's failure
+// path, so a partially-initialized engine is cleaned up the same way; every
+// step is nil-guarded. It does not wait on shutdownWg — the caller does that
+// after releasing the lock, since the goroutines also take syncMsgMux.
+func (e *Engine) stopLocked() {
 	if e.connMgr != nil {
 		e.connMgr.Close()
 	}
@@ -367,10 +408,6 @@ func (e *Engine) Stop() error {
 	// so dbus and friends don't complain because of a missing interface
 	e.stopDNSServer()
 
-	if e.cancel != nil {
-		e.cancel()
-	}
-
 	e.jobExecutorWG.Wait() // block until job goroutines finish
 
 	e.close()
@@ -389,21 +426,6 @@ func (e *Engine) Stop() error {
 	if err := e.stateManager.PersistState(context.Background()); err != nil {
 		log.Errorf("failed to persist state: %v", err)
 	}
-
-	e.syncMsgMux.Unlock()
-
-	timeout := e.calculateShutdownTimeout()
-	log.Debugf("waiting for goroutines to finish with timeout: %v", timeout)
-	shutdownCtx, cancel := context.WithTimeout(context.Background(), timeout)
-	defer cancel()
-
-	if err := waitWithContext(shutdownCtx, &e.shutdownWg); err != nil {
-		log.Warnf("shutdown timeout exceeded after %v, some goroutines may still be running", timeout)
-	}
-
-	log.Infof("stopped Netbird Engine")
-
-	return nil
 }
 
 // calculateShutdownTimeout returns shutdown timeout: 10s base + 100ms per peer, capped at 30s.
@@ -441,18 +463,38 @@ func waitWithContext(ctx context.Context, wg *sync.WaitGroup) error {
 // Start creates a new WireGuard tunnel interface and listens to events from Signal and Management services
 // Connections to remote peers are not established here.
 // However, they will be established once an event with a list of peers to connect to will be received from Management Service
-func (e *Engine) Start(netbirdConfig *mgmProto.NetbirdConfig, mgmtURL *url.URL) error {
+func (e *Engine) Start(netbirdConfig *mgmProto.NetbirdConfig, mgmtURL *url.URL) (err error) {
 	e.syncMsgMux.Lock()
 	defer e.syncMsgMux.Unlock()
 
-	if err := iface.ValidateMTU(e.config.MTU); err != nil {
+	// The engine is single-use. Reject a duplicate start and a start on an
+	// already-stopped engine (run context cancelled).
+	if e.started {
+		return ErrEngineAlreadyStarted
+	}
+
+	if ctxErr := e.ctx.Err(); ctxErr != nil {
+		return fmt.Errorf("engine already stopped: %w", ctxErr)
+	}
+
+	e.started = true
+
+	// Tear down any partially-initialized state on a failed start. Cancel the
+	// run context first so goroutines started before the failure (connMgr,
+	// srWatcher, monitors) unwind, then stopLocked mirrors Stop's teardown (we
+	// already hold syncMsgMux), cleaning up route/DNS/flow/state managers too,
+	// not just what close() covers.
+	defer func() {
+		if err != nil {
+			e.cancel()
+			e.stopLocked()
+		}
+	}()
+
+	if err = iface.ValidateMTU(e.config.MTU); err != nil {
 		return fmt.Errorf("invalid MTU configuration: %w", err)
 	}
 
-	if e.cancel != nil {
-		e.cancel()
-	}
-	e.ctx, e.cancel = context.WithCancel(e.clientCtx)
 	e.exposeManager = expose.NewManager(e.ctx, e.mgmClient)
 
 	wgIface, err := e.newWgIface()
@@ -486,13 +528,11 @@ func (e *Engine) Start(netbirdConfig *mgmProto.NetbirdConfig, mgmtURL *url.URL)
 
 	initialRoutes, dnsConfig, dnsFeatureFlag, err := e.readInitialSettings()
 	if err != nil {
-		e.close()
 		return fmt.Errorf("read initial settings: %w", err)
 	}
 
 	dnsServer, err := e.newDnsServer(dnsConfig)
 	if err != nil {
-		e.close()
 		return fmt.Errorf("create dns server: %w", err)
 	}
 	e.dnsServer = dnsServer
@@ -527,12 +567,14 @@ func (e *Engine) Start(netbirdConfig *mgmProto.NetbirdConfig, mgmtURL *url.URL)
 
 	if err = e.wgInterfaceCreate(); err != nil {
 		log.Errorf("failed creating tunnel interface %s: [%s]", e.config.WgIfaceName, err.Error())
-		e.close()
 		return fmt.Errorf("create wg interface: %w", err)
 	}
 
+	if filteredDevice := e.wgInterface.GetDevice(); filteredDevice != nil {
+		filteredDevice.SetPanicHandler(e.triggerClientRestart)
+	}
+
 	if err := e.createFirewall(); err != nil {
-		e.close()
 		return err
 	}
 
@@ -544,7 +586,6 @@ func (e *Engine) Start(netbirdConfig *mgmProto.NetbirdConfig, mgmtURL *url.URL)
 	e.udpMux, err = e.wgInterface.Up()
 	if err != nil {
 		log.Errorf("failed to pull up wgInterface [%s]: %s", e.wgInterface.Name(), err.Error())
-		e.close()
 		return fmt.Errorf("up wg interface: %w", err)
 	}
 
@@ -569,9 +610,7 @@ func (e *Engine) Start(netbirdConfig *mgmProto.NetbirdConfig, mgmtURL *url.URL)
 		e.acl = acl.NewDefaultManager(e.firewall)
 	}
 
-	err = e.dnsServer.Initialize()
+	if err := e.dnsServer.Initialize(); err != nil {
-	if err != nil {
-		e.close()
 		return fmt.Errorf("initialize dns server: %w", err)
 	}
 
@@ -583,7 +622,9 @@ func (e *Engine) Start(netbirdConfig *mgmProto.NetbirdConfig, mgmtURL *url.URL)
 	e.srWatcher = guard.NewSRWatcher(e.signal, e.relayManager, e.mobileDep.IFaceDiscover, iceCfg)
 	e.srWatcher.Start(peer.IsForceRelayed())
 
-	e.receiveSignalEvents()
+	if err = e.receiveSignalEvents(); err != nil {
+		return err
+	}
 	e.receiveManagementEvents()
 	e.receiveJobEvents()
 
@@ -635,7 +676,6 @@ func (e *Engine) createFirewall() error {
 
 func (e *Engine) initFirewall() error {
 	if err := e.routeManager.SetFirewall(e.firewall); err != nil {
-		e.close()
 		return fmt.Errorf("set firewall: %w", err)
 	}
 
@@ -728,7 +768,15 @@ func (e *Engine) blockLanAccess() {
 
 // modifyPeers updates peers that have been modified (e.g. IP address has been changed).
 // It closes the existing connection, removes it from the peerConns map, and creates a new one.
-func (e *Engine) modifyPeers(peersUpdate []*mgmProto.RemotePeerConfig) error {
+// maxPeersPerSyncPass is the default per-pass cap on how many peers each of
+// removePeers/modifyPeers/addNewPeers applies, so syncMsgMux is held only for a
+// batch at a time and other subsystems can interleave between passes. It is
+// passed in (not read globally) so tests can exercise the multi-pass path.
+const maxPeersPerSyncPass = 300
+
+// modifyPeers re-applies up to maxBatch changed peers per call. It returns true
+// when more changed peers remained than the cap, so the caller re-runs.
+func (e *Engine) modifyPeers(peersUpdate []*mgmProto.RemotePeerConfig, maxBatch int) (bool, error) {
 
 	// first, check if peers have been modified
 	var modified []*mgmProto.RemotePeerConfig
@@ -758,26 +806,32 @@ func (e *Engine) modifyPeers(peersUpdate []*mgmProto.RemotePeerConfig) error {
 		}
 	}
 
+	more := false
+	if len(modified) > maxBatch {
+		modified = modified[:maxBatch]
+		more = true
+	}
+
 	// second, close all modified connections and remove them from the state map
 	for _, p := range modified {
-		err := e.removePeer(p.GetWgPubKey())
+		if err := e.removePeer(p.GetWgPubKey()); err != nil {
-		if err != nil {
+			return false, err
-			return err
 		}
 	}
 	// third, add the peer connections again
 	for _, p := range modified {
-		err := e.addNewPeer(p)
+		if err := e.addNewPeer(p); err != nil {
-		if err != nil {
+			return false, err
-			return err
 		}
 	}
-	return nil
+	return more, nil
 }
 
 // removePeers finds and removes peers that do not exist anymore in the network map received from the Management Service.
 // It also removes peers that have been modified (e.g. change of IP address). They will be added again in addPeers method.
-func (e *Engine) removePeers(peersUpdate []*mgmProto.RemotePeerConfig) error {
+// removePeers removes up to maxBatch peers per call. It returns true when more
+// peers remained to remove than the cap, so the caller re-runs.
+func (e *Engine) removePeers(peersUpdate []*mgmProto.RemotePeerConfig, maxBatch int) (bool, error) {
 	newPeers := make([]string, 0, len(peersUpdate))
 	for _, p := range peersUpdate {
 		newPeers = append(newPeers, p.GetWgPubKey())
@@ -785,14 +839,19 @@ func (e *Engine) removePeers(peersUpdate []*mgmProto.RemotePeerConfig) error {
 
 	toRemove := util.SliceDiff(e.peerStore.PeersPubKey(), newPeers)
 
+	more := false
+	if len(toRemove) > maxBatch {
+		toRemove = toRemove[:maxBatch]
+		more = true
+	}
+
 	for _, p := range toRemove {
-		err := e.removePeer(p)
+		if err := e.removePeer(p); err != nil {
-		if err != nil {
+			return false, err
-			return err
 		}
 		log.Infof("removed peer %s", p)
 	}
-	return nil
+	return more, nil
 }
 
 func (e *Engine) removeAllPeers() error {
@@ -861,92 +920,117 @@ func (e *Engine) handleAutoUpdateVersion(autoUpdateSettings *mgmProto.AutoUpdate
 	e.updateManager.SetVersion(autoUpdateSettings.Version, autoUpdateSettings.AlwaysUpdate)
 }
 
-func (e *Engine) handleSync(update *mgmProto.SyncResponse) error {
+// applySyncPass applies one bounded pass of the sync update under syncMsgMux and
-	started := time.Now()
+// returns true if more peers remained than the per-pass cap. It is driven by the
-	defer func() {
+// mapStateManager, which re-invokes it (releasing the lock between passes) until
-		duration := time.Since(started)
+// the update is fully applied.
-		log.Infof("sync finished in %s", duration)
+func (e *Engine) applySyncPass(update *mgmProto.SyncResponse, firstPass bool) (bool, error) {
-		e.clientMetrics.RecordSyncDuration(e.ctx, duration)
-	}()
 	e.syncMsgMux.Lock()
 	defer e.syncMsgMux.Unlock()
 
 	// Check context INSIDE lock to ensure atomicity with shutdown
 	if e.ctx.Err() != nil {
-		return e.ctx.Err()
+		return false, e.ctx.Err()
 	}
 
 	if update.NetworkMap != nil && update.NetworkMap.PeerConfig != nil {
 		e.handleAutoUpdateVersion(update.NetworkMap.PeerConfig.AutoUpdate)
 	}
 
-	if update.GetNetbirdConfig() != nil {
+	if err := e.updateNetbirdConfig(update.GetNetbirdConfig()); err != nil {
-		wCfg := update.GetNetbirdConfig()
+		return false, err
-		err := e.updateTURNs(wCfg.GetTurns())
+	}
-		if err != nil {
-			return fmt.Errorf("update TURNs: %w", err)
-		}
 
-		err = e.updateSTUNs(wCfg.GetStuns())
+	// Posture checks are bound to the network map presence:
-		if err != nil {
+	//   NetworkMap != nil, checks present -> apply the received checks
-			return fmt.Errorf("update STUNs: %w", err)
+	//   NetworkMap != nil, checks nil     -> posture checks were removed, clear them
-		}
+	//   NetworkMap == nil                 -> config-only update (e.g. relay token rotation),
-
+	//                                        leave the previously applied checks untouched
-		var stunTurn []*stun.URI
+	nm := update.GetNetworkMap()
-		stunTurn = append(stunTurn, e.STUNs...)
+	if nm == nil {
-		stunTurn = append(stunTurn, e.TURNs...)
+		return false, nil
-		e.stunTurn.Store(stunTurn)
-
-		err = e.handleRelayUpdate(wCfg.GetRelay())
-		if err != nil {
-			return err
-		}
-
-		err = e.handleFlowUpdate(wCfg.GetFlow())
-		if err != nil {
-			return fmt.Errorf("handle the flow configuration: %w", err)
-		}
-
-		if err := e.PopulateNetbirdConfig(wCfg, nil); err != nil {
-			log.Warnf("Failed to update DNS server config: %v", err)
-		}
-
-		// todo update signal
 	}
 
 	if err := e.updateChecksIfNew(update.Checks); err != nil {
-		return err
+		return false, err
 	}
 
-	nm := update.GetNetworkMap()
-	if nm == nil {
-		return nil
-	}
-
-	// Persist sync response under the dedicated lock (syncRespMux), not under syncMsgMux.
-	// A non-nil syncStore is what marks persistence as enabled. Hold the lock for
-	// the whole Set so the store cannot be cleared (disabled / engine close)
-	// mid-call and have this write resurrect a file that was just removed.
-	e.syncRespMux.RLock()
-	if e.syncStore != nil {
-		if err := e.syncStore.Set(update); err != nil {
-			log.Errorf("failed to persist sync response: %v", err)
-		} else {
-			log.Debugf("sync response persisted with serial %d", nm.GetSerial())
-		}
-	}
-	e.syncRespMux.RUnlock()
-
 	// only apply new changes and ignore old ones
-	if err := e.updateNetworkMap(nm); err != nil {
+	more, err := e.updateNetworkMap(nm, maxPeersPerSyncPass, firstPass)
-		return err
+	if err != nil {
+		return false, err
 	}
 
 	e.statusRecorder.PublishEvent(cProto.SystemEvent_INFO, cProto.SystemEvent_SYSTEM, "Network map updated", "", nil)
 
+	return more, nil
+}
+
+// updateNetbirdConfig applies the management-provided NetBird configuration:
+// STUN/TURN and relay servers, flow logging and DNS settings. A nil config is a no-op,
+// which is the case for sync updates carrying only a network map.
+func (e *Engine) updateNetbirdConfig(wCfg *mgmProto.NetbirdConfig) error {
+	if wCfg == nil {
+		return nil
+	}
+
+	if err := e.updateTURNs(wCfg.GetTurns()); err != nil {
+		return fmt.Errorf("update TURNs: %w", err)
+	}
+
+	if err := e.updateSTUNs(wCfg.GetStuns()); err != nil {
+		return fmt.Errorf("update STUNs: %w", err)
+	}
+
+	var stunTurn []*stun.URI
+	stunTurn = append(stunTurn, e.STUNs...)
+	stunTurn = append(stunTurn, e.TURNs...)
+	e.stunTurn.Store(stunTurn)
+
+	if err := e.handleRelayUpdate(wCfg.GetRelay()); err != nil {
+		return err
+	}
+
+	if err := e.handleFlowUpdate(wCfg.GetFlow()); err != nil {
+		return fmt.Errorf("handle the flow configuration: %w", err)
+	}
+
+	if err := e.PopulateNetbirdConfig(wCfg, nil); err != nil {
+		log.Warnf("Failed to update DNS server config: %v", err)
+	}
+
+	// todo update signal
+
 	return nil
 }
 
+// persistSyncResponse stores the full sync response so it can be restored on the next
+// startup. Persistence is enabled only when syncStore is set. The dedicated syncRespMux
+// (not syncMsgMux) is held for the whole Set so the store cannot be cleared (disabled /
+// engine close) mid-call and have this write resurrect a file that was just removed.
+func (e *Engine) persistSyncResponse(update *mgmProto.SyncResponse) {
+	// Only persist updates that carry a network map. Config-only updates (e.g. relay
+	// token rotation, STUN/TURN) have a nil NetworkMap; persisting them would overwrite
+	// the last full map on disk and break restore-on-restart.
+	if update.GetNetworkMap() == nil {
+		return
+	}
+
+	e.syncRespMux.RLock()
+	defer e.syncRespMux.RUnlock()
+
+	if e.syncStore == nil {
+		return
+	}
+
+	if err := e.syncStore.Set(update); err != nil {
+		log.Errorf("failed to persist sync response: %v", err)
+		return
+	}
+
+	log.Debugf("sync response persisted with serial %d", update.GetNetworkMap().GetSerial())
+}
+
 func (e *Engine) handleRelayUpdate(update *mgmProto.RelayConfig) error {
 	if update != nil {
 		// when we receive token we expect valid address list too
@@ -1011,7 +1095,7 @@ func (e *Engine) updateChecksIfNew(checks []*mgmProto.Checks) error {
 	}
 	e.checks = checks
 
-	info, err := system.GetInfoWithChecks(e.ctx, checks)
+	info, err := system.GetInfoWithChecks(e.ctx, checks, e.overlayAddresses()...)
 	if err != nil {
 		log.Warnf("failed to get system info with checks: %v", err)
 		info = system.GetInfo(e.ctx)
@@ -1042,6 +1126,20 @@ func (e *Engine) updateChecksIfNew(checks []*mgmProto.Checks) error {
 	return nil
 }
 
+// overlayAddresses returns our own WireGuard overlay address (v4 and v6) so it
+// can be excluded from the reported network addresses; the interface coming and
+// going otherwise churns the peer meta on the management server.
+func (e *Engine) overlayAddresses() []netip.Addr {
+	var ips []netip.Addr
+	if e.config.WgAddr.IP.IsValid() {
+		ips = append(ips, e.config.WgAddr.IP)
+	}
+	if e.config.WgAddr.HasIPv6() {
+		ips = append(ips, e.config.WgAddr.IPv6)
+	}
+	return ips
+}
+
 func (e *Engine) updateConfig(conf *mgmProto.PeerConfig) error {
 	if e.wgInterface == nil {
 		return errors.New("wireguard interface is not initialized")
@@ -1185,7 +1283,7 @@ func (e *Engine) receiveManagementEvents() {
 	e.shutdownWg.Add(1)
 	go func() {
 		defer e.shutdownWg.Done()
-		info, err := system.GetInfoWithChecks(e.ctx, e.checks)
+		info, err := system.GetInfoWithChecks(e.ctx, e.checks, e.overlayAddresses()...)
 		if err != nil {
 			log.Warnf("failed to get system info with checks: %v", err)
 			info = system.GetInfo(e.ctx)
@@ -1209,7 +1307,19 @@ func (e *Engine) receiveManagementEvents() {
 			e.config.DisableSSHAuth,
 		)
 
-		err = e.mgmClient.Sync(e.ctx, info, e.handleSync)
+		// The map-state manager converges the latest update in the background in
+		// bounded passes; the stream callback only hands it the newest target.
+		manager := newMapStateManager(e.applySyncPass, e.persistSyncResponse, func(d time.Duration) {
+			log.Infof("sync finished in %s", d)
+			e.clientMetrics.RecordSyncDuration(e.ctx, d)
+		})
+		e.shutdownWg.Add(1)
+		go func() {
+			defer e.shutdownWg.Done()
+			manager.run(e.ctx)
+		}()
+
+		err = e.mgmClient.Sync(e.ctx, info, manager.SetTarget)
 		if err != nil {
 			// happens if management is unavailable for a long time.
 			// We want to cancel the operation of the whole client
@@ -1260,21 +1370,104 @@ func (e *Engine) updateTURNs(turns []*mgmProto.ProtectedHostConfig) error {
 	return nil
 }
 
-func (e *Engine) updateNetworkMap(networkMap *mgmProto.NetworkMap) error {
+// updateNetworkMap applies the wholesale parts (config, routes, ACL, DNS) in full
+// and up to maxBatch peers per phase. It returns true when more peers remained
+// than the cap, so the caller re-runs until convergence.
+func (e *Engine) updateNetworkMap(networkMap *mgmProto.NetworkMap, maxBatch int, firstPass bool) (bool, error) {
 	// intentionally leave it before checking serial because for now it can happen that peer IP changed but serial didn't
 	if networkMap.GetPeerConfig() != nil {
 		err := e.updateConfig(networkMap.GetPeerConfig())
 		if err != nil {
-			return err
+			return false, err
 		}
 	}
 
 	serial := networkMap.GetSerial()
 	if e.networkSerial > serial {
 		log.Debugf("received outdated NetworkMap with serial %d, ignoring", serial)
-		return nil
+		return false, nil
 	}
 
+	// Wholesale sections (firewall/ACL, DNS, routes, forward rules) are applied
+	// up-front and only once per target: they are cheap, local, idempotent and must
+	// be in place before peers come up (fail-closed). On the bounded re-runs that only
+	// drain the remaining peer batches they are skipped — the applied forward rules are
+	// reused from e.forwardingRules for the lazy-exclude finalize.
+	if firstPass {
+		e.applyWholesale(networkMap, serial)
+	}
+
+	log.Debugf("got peers update from Management Service, total peers to connect to = %d", len(networkMap.GetRemotePeers()))
+
+	e.updateOfflinePeers(networkMap.GetOfflinePeers())
+
+	// Filter out own peer from the remote peers list
+	localPubKey := e.config.WgPrivateKey.PublicKey().String()
+	remotePeers := make([]*mgmProto.RemotePeerConfig, 0, len(networkMap.GetRemotePeers()))
+	for _, p := range networkMap.GetRemotePeers() {
+		if p.GetWgPubKey() != localPubKey {
+			remotePeers = append(remotePeers, p)
+		}
+	}
+
+	// needMore signals the caller to re-run when a peer phase hit its per-pass cap.
+	needMore := false
+
+	// cleanup request, most likely our peer has been deleted
+	if networkMap.GetRemotePeersIsEmpty() {
+		err := e.removeAllPeers()
+		e.statusRecorder.FinishPeerListModifications()
+		if err != nil {
+			return false, err
+		}
+	} else {
+		removeMore, err := e.removePeers(remotePeers, maxBatch)
+		if err != nil {
+			return false, err
+		}
+
+		modifyMore, err := e.modifyPeers(remotePeers, maxBatch)
+		if err != nil {
+			return false, err
+		}
+
+		addMore, err := e.addNewPeers(remotePeers, maxBatch)
+		if err != nil {
+			return false, err
+		}
+
+		needMore = removeMore || modifyMore || addMore
+
+		e.statusRecorder.FinishPeerListModifications()
+
+		e.updatePeerSSHHostKeys(remotePeers)
+
+		if err := e.updateSSHClientConfig(remotePeers); err != nil {
+			log.Warnf("failed to update SSH client config: %v", err)
+		}
+
+		e.updateSSHServerAuth(networkMap.GetSshAuth())
+	}
+
+	// Set the exclude list only once peers have fully converged (this pass added
+	// the last batch). It needs all target peers present in the store, and
+	// ExcludePeer has replace-semantics — a partial set mid-convergence would be wrong.
+	if !needMore {
+		excludedLazyPeers := e.toExcludedLazyPeers(e.forwardingRules, remotePeers)
+		e.connMgr.SetExcludeList(e.ctx, excludedLazyPeers)
+	}
+
+	e.networkSerial = serial
+
+	return needMore, nil
+}
+
+// applyWholesale applies the cheap, local, idempotent map sections — lazy feature
+// flag, firewall/legacy management, DNS, routes, ACL filtering, DNS forwarder and
+// ingress forward rules — that must be in place before peers come up. It runs once
+// per target (first pass only); the resulting forward rules are stashed in
+// e.forwardingRules for the lazy-exclude finalize on the peer-converged pass.
+func (e *Engine) applyWholesale(networkMap *mgmProto.NetworkMap, serial uint64) {
 	if err := e.connMgr.UpdatedRemoteFeatureFlag(e.ctx, networkMap.GetPeerConfig().GetLazyConnectionEnabled()); err != nil {
 		log.Errorf("failed to update lazy connection feature flag: %v", err)
 	}
@@ -1335,61 +1528,7 @@ func (e *Engine) updateNetworkMap(networkMap *mgmProto.NetworkMap) error {
 	if err != nil {
 		log.Errorf("failed to update forward rules, err: %v", err)
 	}
-
+	e.forwardingRules = forwardingRules
-	log.Debugf("got peers update from Management Service, total peers to connect to = %d", len(networkMap.GetRemotePeers()))
-
-	e.updateOfflinePeers(networkMap.GetOfflinePeers())
-
-	// Filter out own peer from the remote peers list
-	localPubKey := e.config.WgPrivateKey.PublicKey().String()
-	remotePeers := make([]*mgmProto.RemotePeerConfig, 0, len(networkMap.GetRemotePeers()))
-	for _, p := range networkMap.GetRemotePeers() {
-		if p.GetWgPubKey() != localPubKey {
-			remotePeers = append(remotePeers, p)
-		}
-	}
-
-	// cleanup request, most likely our peer has been deleted
-	if networkMap.GetRemotePeersIsEmpty() {
-		err := e.removeAllPeers()
-		e.statusRecorder.FinishPeerListModifications()
-		if err != nil {
-			return err
-		}
-	} else {
-		err := e.removePeers(remotePeers)
-		if err != nil {
-			return err
-		}
-
-		err = e.modifyPeers(remotePeers)
-		if err != nil {
-			return err
-		}
-
-		err = e.addNewPeers(remotePeers)
-		if err != nil {
-			return err
-		}
-
-		e.statusRecorder.FinishPeerListModifications()
-
-		e.updatePeerSSHHostKeys(remotePeers)
-
-		if err := e.updateSSHClientConfig(remotePeers); err != nil {
-			log.Warnf("failed to update SSH client config: %v", err)
-		}
-
-		e.updateSSHServerAuth(networkMap.GetSshAuth())
-	}
-
-	// must set the exclude list after the peers are added. Without it the manager can not figure out the peers parameters from the store
-	excludedLazyPeers := e.toExcludedLazyPeers(forwardingRules, remotePeers)
-	e.connMgr.SetExcludeList(e.ctx, excludedLazyPeers)
-
-	e.networkSerial = serial
-
-	return nil
 }
 
 func toDNSFeatureFlag(networkMap *mgmProto.NetworkMap) bool {
@@ -1569,14 +1708,23 @@ func addrToString(addr netip.Addr) string {
 }
 
 // addNewPeers adds peers that were not know before but arrived from the Management service with the update
-func (e *Engine) addNewPeers(peersUpdate []*mgmProto.RemotePeerConfig) error {
+// addNewPeers adds up to maxBatch not-yet-present peers per call. It returns true
+// when more new peers remained than the cap, so the caller re-runs.
+func (e *Engine) addNewPeers(peersUpdate []*mgmProto.RemotePeerConfig, maxBatch int) (bool, error) {
+	added := 0
 	for _, p := range peersUpdate {
-		err := e.addNewPeer(p)
+		if _, ok := e.peerStore.PeerConn(p.GetWgPubKey()); ok {
-		if err != nil {
+			continue // already present (cheap skip), does not count toward the cap
-			return err
 		}
+		if added >= maxBatch {
+			return true, nil // at least one more new peer remains
+		}
+		if err := e.addNewPeer(p); err != nil {
+			return false, err
+		}
+		added++
 	}
-	return nil
+	return false, nil
 }
 
 // addNewPeer add peer if connection doesn't exist
@@ -1674,7 +1822,7 @@ func (e *Engine) createPeerConn(pubKey string, allowedIPs []netip.Prefix, agentV
 }
 
 // receiveSignalEvents connects to the Signal Service event stream to negotiate connection with remote peers
-func (e *Engine) receiveSignalEvents() {
+func (e *Engine) receiveSignalEvents() error {
 	e.shutdownWg.Add(1)
 	go func() {
 		defer e.shutdownWg.Done()
@@ -1690,6 +1838,13 @@ func (e *Engine) receiveSignalEvents() {
 				return e.ctx.Err()
 			}
 
+			// Self-addressed heartbeat: the signal client's receive watchdog
+			// round-trips this through the server to confirm the receive stream
+			// is delivering. Liveness is already recorded before this handler.
+			if msg.GetBody().GetType() == sProto.Body_HEARTBEAT {
+				return nil
+			}
+
 			conn, ok := e.peerStore.PeerConn(msg.Key)
 			if !ok {
 				return fmt.Errorf("wrongly addressed message %s", msg.Key)
@@ -1738,7 +1893,12 @@ func (e *Engine) receiveSignalEvents() {
 		}
 	}()
 
-	e.signal.WaitStreamConnected()
+	// todo: consider to remove this blocker. I do not see benefit to block the Start operations
+	e.signal.WaitStreamConnected(e.ctx)
+	if err := e.ctx.Err(); err != nil {
+		return fmt.Errorf("wait for signal stream: %w", err)
+	}
+	return nil
 }
 
 func (e *Engine) parseNATExternalIPMappings() []string {
@@ -1888,7 +2048,6 @@ func (e *Engine) newWgIface() (*iface.WGIface, error) {
 		WGPrivKey:    e.config.WgPrivateKey.String(),
 		MTU:          e.config.MTU,
 		TransportNet: transportNet,
-		FilterFn:     e.addrViaRoutes,
 		DisableDNS:   e.config.DisableDNS,
 	}
 
@@ -2136,21 +2295,6 @@ func (e *Engine) startNetworkMonitor() {
 	}()
 }
 
-func (e *Engine) addrViaRoutes(addr netip.Addr) (bool, netip.Prefix, error) {
-	var vpnRoutes []netip.Prefix
-	for _, routes := range e.routeManager.GetClientRoutes() {
-		if len(routes) > 0 && routes[0] != nil {
-			vpnRoutes = append(vpnRoutes, routes[0].Network)
-		}
-	}
-
-	if isVpn, prefix := systemops.IsAddrRouted(addr, vpnRoutes); isVpn {
-		return true, prefix, nil
-	}
-
-	return false, netip.Prefix{}, nil
-}
-
 func (e *Engine) stopDNSServer() {
 	if e.dnsServer == nil {
 		return

client/internal/engine_privileged_test.go

@@ -0,0 +1,565 @@
+//go:build privileged
+
+package internal
+
+import (
+	"context"
+	"fmt"
+	"net"
+	"runtime"
+	"strings"
+	"sync"
+	"testing"
+	"time"
+
+	"github.com/golang/mock/gomock"
+	"github.com/google/uuid"
+	log "github.com/sirupsen/logrus"
+	"github.com/stretchr/testify/assert"
+	"github.com/stretchr/testify/require"
+	"go.opentelemetry.io/otel"
+	"golang.zx2c4.com/wireguard/wgctrl/wgtypes"
+	"google.golang.org/grpc"
+	"google.golang.org/grpc/keepalive"
+
+	"github.com/netbirdio/netbird/client/iface"
+	"github.com/netbirdio/netbird/client/iface/device"
+	"github.com/netbirdio/netbird/client/iface/wgaddr"
+	"github.com/netbirdio/netbird/client/internal/dns"
+	"github.com/netbirdio/netbird/client/internal/peer"
+	nbssh "github.com/netbirdio/netbird/client/ssh"
+	"github.com/netbirdio/netbird/client/system"
+	nbdns "github.com/netbirdio/netbird/dns"
+	"github.com/netbirdio/netbird/management/internals/controllers/network_map/controller"
+	"github.com/netbirdio/netbird/management/internals/controllers/network_map/update_channel"
+	"github.com/netbirdio/netbird/management/internals/modules/peers"
+	"github.com/netbirdio/netbird/management/internals/modules/peers/ephemeral/manager"
+	"github.com/netbirdio/netbird/management/internals/server/config"
+	nbgrpc "github.com/netbirdio/netbird/management/internals/shared/grpc"
+	"github.com/netbirdio/netbird/management/server"
+	"github.com/netbirdio/netbird/management/server/activity"
+	nbcache "github.com/netbirdio/netbird/management/server/cache"
+	"github.com/netbirdio/netbird/management/server/groups"
+	"github.com/netbirdio/netbird/management/server/integrations/integrated_validator/validator"
+	"github.com/netbirdio/netbird/management/server/integrations/port_forwarding"
+	"github.com/netbirdio/netbird/management/server/job"
+	"github.com/netbirdio/netbird/management/server/permissions"
+	"github.com/netbirdio/netbird/management/server/settings"
+	"github.com/netbirdio/netbird/management/server/store"
+	"github.com/netbirdio/netbird/management/server/telemetry"
+	"github.com/netbirdio/netbird/management/server/types"
+	mgmt "github.com/netbirdio/netbird/shared/management/client"
+	mgmtProto "github.com/netbirdio/netbird/shared/management/proto"
+	relayClient "github.com/netbirdio/netbird/shared/relay/client"
+	signal "github.com/netbirdio/netbird/shared/signal/client"
+	"github.com/netbirdio/netbird/shared/signal/proto"
+	signalServer "github.com/netbirdio/netbird/signal/server"
+	"github.com/netbirdio/netbird/util"
+)
+
+func TestEngine_SSH(t *testing.T) {
+	key, err := wgtypes.GeneratePrivateKey()
+	if err != nil {
+		t.Fatal(err)
+		return
+	}
+
+	sshKey, err := nbssh.GeneratePrivateKey(nbssh.ED25519)
+	if err != nil {
+		t.Fatal(err)
+		return
+	}
+
+	ctx, cancel := context.WithCancel(CtxInitState(context.Background()))
+	defer cancel()
+
+	relayMgr := relayClient.NewManager(ctx, nil, key.PublicKey().String(), iface.DefaultMTU)
+	engine := NewEngine(
+		ctx, cancel,
+		&EngineConfig{
+			WgIfaceName:      "utun101",
+			WgAddr:           wgaddr.MustParseWGAddress("100.64.0.1/24"),
+			WgPrivateKey:     key,
+			WgPort:           33100,
+			ServerSSHAllowed: true,
+			MTU:              iface.DefaultMTU,
+			SSHKey:           sshKey,
+		},
+		EngineServices{
+			SignalClient:   &signal.MockClient{},
+			MgmClient:      &mgmt.MockClient{},
+			RelayManager:   relayMgr,
+			StatusRecorder: peer.NewRecorder("https://mgm"),
+		},
+		MobileDependency{},
+	)
+
+	engine.dnsServer = &dns.MockServer{
+		UpdateDNSServerFunc: func(serial uint64, update nbdns.Config) error { return nil },
+	}
+
+	err = engine.Start(nil, nil)
+	require.NoError(t, err)
+
+	defer func() {
+		err := engine.Stop()
+		if err != nil {
+			return
+		}
+	}()
+
+	peerWithSSH := &mgmtProto.RemotePeerConfig{
+		WgPubKey:   "MNHf3Ma6z6mdLbriAJbqhX7+nM/B71lgw2+91q3LfhU=",
+		AllowedIps: []string{"100.64.0.21/24"},
+		SshConfig: &mgmtProto.SSHConfig{
+			SshPubKey: []byte("ssh-ed25519 AAAAC3NzaC1lZDI1NTE5AAAAIFATYCqaQw/9id1Qkq3n16JYhDhXraI6Pc1fgB8ynEfQ"),
+		},
+	}
+
+	// SSH server is not enabled so SSH config of a remote peer should be ignored
+	networkMap := &mgmtProto.NetworkMap{
+		Serial:             6,
+		PeerConfig:         nil,
+		RemotePeers:        []*mgmtProto.RemotePeerConfig{peerWithSSH},
+		RemotePeersIsEmpty: false,
+	}
+
+	_, err = engine.updateNetworkMap(networkMap, maxPeersPerSyncPass, true)
+	require.NoError(t, err)
+
+	assert.Nil(t, engine.sshServer)
+
+	// SSH server is enabled, therefore SSH config should be applied
+	networkMap = &mgmtProto.NetworkMap{
+		Serial: 7,
+		PeerConfig: &mgmtProto.PeerConfig{Address: "100.64.0.1/24",
+			SshConfig: &mgmtProto.SSHConfig{
+				SshEnabled: true,
+				JwtConfig: &mgmtProto.JWTConfig{
+					Issuer:       "test-issuer",
+					Audience:     "test-audience",
+					KeysLocation: "test-keys",
+					MaxTokenAge:  3600,
+				},
+			}},
+		RemotePeers:        []*mgmtProto.RemotePeerConfig{peerWithSSH},
+		RemotePeersIsEmpty: false,
+	}
+
+	_, err = engine.updateNetworkMap(networkMap, maxPeersPerSyncPass, true)
+	require.NoError(t, err)
+
+	time.Sleep(250 * time.Millisecond)
+	assert.NotNil(t, engine.sshServer)
+
+	// now remove peer
+	networkMap = &mgmtProto.NetworkMap{
+		Serial:             8,
+		RemotePeers:        []*mgmtProto.RemotePeerConfig{},
+		RemotePeersIsEmpty: false,
+	}
+
+	_, err = engine.updateNetworkMap(networkMap, maxPeersPerSyncPass, true)
+	require.NoError(t, err)
+
+	// time.Sleep(250 * time.Millisecond)
+	assert.NotNil(t, engine.sshServer)
+
+	// now disable SSH server
+	networkMap = &mgmtProto.NetworkMap{
+		Serial: 9,
+		PeerConfig: &mgmtProto.PeerConfig{Address: "100.64.0.1/24",
+			SshConfig: &mgmtProto.SSHConfig{SshEnabled: false}},
+		RemotePeers:        []*mgmtProto.RemotePeerConfig{peerWithSSH},
+		RemotePeersIsEmpty: false,
+	}
+
+	_, err = engine.updateNetworkMap(networkMap, maxPeersPerSyncPass, true)
+	require.NoError(t, err)
+
+	assert.Nil(t, engine.sshServer)
+}
+
+func TestEngine_Sync(t *testing.T) {
+	key, err := wgtypes.GeneratePrivateKey()
+	if err != nil {
+		t.Fatal(err)
+		return
+	}
+
+	ctx, cancel := context.WithCancel(CtxInitState(context.Background()))
+	defer cancel()
+
+	// feed updates to Engine via mocked Management client
+	updates := make(chan *mgmtProto.SyncResponse)
+	defer close(updates)
+	syncFunc := func(ctx context.Context, info *system.Info, msgHandler func(msg *mgmtProto.SyncResponse) error) error {
+		for msg := range updates {
+			err := msgHandler(msg)
+			if err != nil {
+				t.Fatal(err)
+			}
+		}
+		return nil
+	}
+	relayMgr := relayClient.NewManager(ctx, nil, key.PublicKey().String(), iface.DefaultMTU)
+	engine := NewEngine(ctx, cancel, &EngineConfig{
+		WgIfaceName:  "utun103",
+		WgAddr:       wgaddr.MustParseWGAddress("100.64.0.1/24"),
+		WgPrivateKey: key,
+		WgPort:       33100,
+		MTU:          iface.DefaultMTU,
+	}, EngineServices{
+		SignalClient:   &signal.MockClient{},
+		MgmClient:      &mgmt.MockClient{SyncFunc: syncFunc},
+		RelayManager:   relayMgr,
+		StatusRecorder: peer.NewRecorder("https://mgm"),
+	}, MobileDependency{})
+	engine.ctx = ctx
+
+	engine.dnsServer = &dns.MockServer{
+		UpdateDNSServerFunc: func(serial uint64, update nbdns.Config) error { return nil },
+	}
+
+	defer func() {
+		err := engine.Stop()
+		if err != nil {
+			return
+		}
+	}()
+
+	err = engine.Start(nil, nil)
+	if err != nil {
+		t.Fatal(err)
+		return
+	}
+
+	peer1 := &mgmtProto.RemotePeerConfig{
+		WgPubKey:   "RRHf3Ma6z6mdLbriAJbqhX7+nM/B71lgw2+91q3LfhU=",
+		AllowedIps: []string{"100.64.0.10/24"},
+	}
+	peer2 := &mgmtProto.RemotePeerConfig{
+		WgPubKey:   "LLHf3Ma6z6mdLbriAJbqhX9+nM/B71lgw2+91q3LlhU=",
+		AllowedIps: []string{"100.64.0.11/24"},
+	}
+	peer3 := &mgmtProto.RemotePeerConfig{
+		WgPubKey:   "GGHf3Ma6z6mdLbriAJbqhX9+nM/B71lgw2+91q3LlhU=",
+		AllowedIps: []string{"100.64.0.12/24"},
+	}
+	// 1st update with just 1 peer and serial larger than the current serial of the engine => apply update
+	updates <- &mgmtProto.SyncResponse{
+		NetworkMap: &mgmtProto.NetworkMap{
+			Serial:             10,
+			PeerConfig:         nil,
+			RemotePeers:        []*mgmtProto.RemotePeerConfig{peer1, peer2, peer3},
+			RemotePeersIsEmpty: false,
+		},
+	}
+
+	timeout := time.After(time.Second * 2)
+	for {
+		select {
+		case <-timeout:
+			t.Fatalf("timeout while waiting for test to finish")
+			return
+		default:
+		}
+
+		if getPeers(engine) == 3 && engine.networkSerial == 10 {
+			break
+		}
+	}
+}
+
+func TestEngine_MultiplePeers(t *testing.T) {
+	// log.SetLevel(log.DebugLevel)
+
+	ctx, cancel := context.WithCancel(CtxInitState(context.Background()))
+	defer cancel()
+
+	sigServer, signalAddr, err := startSignal(t)
+	if err != nil {
+		t.Fatal(err)
+		return
+	}
+	defer sigServer.Stop()
+	mgmtServer, mgmtAddr, err := startManagement(t, t.TempDir(), "../testdata/store.sql")
+	if err != nil {
+		t.Fatal(err)
+		return
+	}
+	defer mgmtServer.GracefulStop()
+
+	setupKey := "A2C8E62B-38F5-4553-B31E-DD66C696CEBB"
+
+	mu := sync.Mutex{}
+	engines := []*Engine{}
+	numPeers := 10
+	wg := sync.WaitGroup{}
+	wg.Add(numPeers)
+	// create and start peers
+	for i := 0; i < numPeers; i++ {
+		j := i
+		go func() {
+			engine, err := createEngine(ctx, cancel, setupKey, j, mgmtAddr, signalAddr)
+			if err != nil {
+				wg.Done()
+				t.Errorf("unable to create the engine for peer %d with error %v", j, err)
+				return
+			}
+			engine.dnsServer = &dns.MockServer{}
+			mu.Lock()
+			defer mu.Unlock()
+			guid := fmt.Sprintf("{%s}", uuid.New().String())
+			device.CustomWindowsGUIDString = strings.ToLower(guid)
+			err = engine.Start(nil, nil)
+			if err != nil {
+				t.Errorf("unable to start engine for peer %d with error %v", j, err)
+				wg.Done()
+				return
+			}
+			engines = append(engines, engine)
+			wg.Done()
+		}()
+	}
+
+	// wait until all have been created and started
+	wg.Wait()
+	if len(engines) != numPeers {
+		t.Fatal("not all peers were started")
+	}
+	// check whether all the peer have expected peers connected
+
+	expectedConnected := numPeers * (numPeers - 1)
+
+	// adjust according to timeouts
+	timeout := 50 * time.Second
+	timeoutChan := time.After(timeout)
+	ticker := time.NewTicker(time.Second)
+	defer ticker.Stop()
+loop:
+	for {
+		select {
+		case <-timeoutChan:
+			t.Fatalf("waiting for expected connections timeout after %s", timeout.String())
+			break loop
+		case <-ticker.C:
+			totalConnected := 0
+			for _, engine := range engines {
+				totalConnected += getConnectedPeers(engine)
+			}
+			if totalConnected == expectedConnected {
+				log.Infof("total connected=%d", totalConnected)
+				break loop
+			}
+			log.Infof("total connected=%d", totalConnected)
+		}
+	}
+	// cleanup test
+	for n, peerEngine := range engines {
+		t.Logf("stopping peer with interface %s from multipeer test, loopIndex %d", peerEngine.wgInterface.Name(), n)
+		errStop := peerEngine.mgmClient.Close()
+		if errStop != nil {
+			log.Infoln("got error trying to close management clients from engine: ", errStop)
+		}
+		errStop = peerEngine.Stop()
+		if errStop != nil {
+			log.Infoln("got error trying to close testing peers engine: ", errStop)
+		}
+	}
+}
+
+var (
+	kaep = keepalive.EnforcementPolicy{
+		MinTime:             15 * time.Second,
+		PermitWithoutStream: true,
+	}
+
+	kasp = keepalive.ServerParameters{
+		MaxConnectionIdle:     15 * time.Second,
+		MaxConnectionAgeGrace: 5 * time.Second,
+		Time:                  5 * time.Second,
+		Timeout:               2 * time.Second,
+	}
+)
+
+func createEngine(ctx context.Context, cancel context.CancelFunc, setupKey string, i int, mgmtAddr string, signalAddr string) (*Engine, error) {
+	key, err := wgtypes.GeneratePrivateKey()
+	if err != nil {
+		return nil, err
+	}
+	mgmtClient, err := mgmt.NewClient(ctx, mgmtAddr, key, false)
+	if err != nil {
+		return nil, err
+	}
+	signalClient, err := signal.NewClient(ctx, signalAddr, key, false)
+	if err != nil {
+		return nil, err
+	}
+
+	info := system.GetInfo(ctx)
+	resp, err := mgmtClient.Register(setupKey, "", info, nil, nil)
+	if err != nil {
+		return nil, err
+	}
+
+	var ifaceName string
+	if runtime.GOOS == "darwin" {
+		ifaceName = fmt.Sprintf("utun1%d", i)
+	} else {
+		ifaceName = fmt.Sprintf("wt%d", i)
+	}
+
+	wgPort := 33100 + i
+	conf := &EngineConfig{
+		WgIfaceName:  ifaceName,
+		WgAddr:       wgaddr.MustParseWGAddress(resp.PeerConfig.Address),
+		WgPrivateKey: key,
+		WgPort:       wgPort,
+		MTU:          iface.DefaultMTU,
+	}
+
+	relayMgr := relayClient.NewManager(ctx, nil, key.PublicKey().String(), iface.DefaultMTU)
+	e, err := NewEngine(ctx, cancel, conf, EngineServices{
+		SignalClient:   signalClient,
+		MgmClient:      mgmtClient,
+		RelayManager:   relayMgr,
+		StatusRecorder: peer.NewRecorder("https://mgm"),
+	}, MobileDependency{}), nil
+	e.ctx = ctx
+	return e, err
+}
+
+func startSignal(t *testing.T) (*grpc.Server, string, error) {
+	t.Helper()
+
+	s := grpc.NewServer(grpc.KeepaliveEnforcementPolicy(kaep), grpc.KeepaliveParams(kasp))
+
+	lis, err := net.Listen("tcp", "localhost:0")
+	if err != nil {
+		log.Fatalf("failed to listen: %v", err)
+	}
+
+	srv, err := signalServer.NewServer(context.Background(), otel.Meter(""))
+	require.NoError(t, err)
+	proto.RegisterSignalExchangeServer(s, srv)
+
+	go func() {
+		if err = s.Serve(lis); err != nil {
+			log.Fatalf("failed to serve: %v", err)
+		}
+	}()
+
+	return s, lis.Addr().String(), nil
+}
+
+func startManagement(t *testing.T, dataDir, testFile string) (*grpc.Server, string, error) {
+	t.Helper()
+
+	config := &config.Config{
+		Stuns:      []*config.Host{},
+		TURNConfig: &config.TURNConfig{},
+		Relay: &config.Relay{
+			Addresses:      []string{"127.0.0.1:1234"},
+			CredentialsTTL: util.Duration{Duration: time.Hour},
+			Secret:         "222222222222222222",
+		},
+		Signal: &config.Host{
+			Proto: "http",
+			URI:   "localhost:10000",
+		},
+		Datadir:    dataDir,
+		HttpConfig: nil,
+	}
+
+	lis, err := net.Listen("tcp", "localhost:0")
+	if err != nil {
+		return nil, "", err
+	}
+	s := grpc.NewServer(grpc.KeepaliveEnforcementPolicy(kaep), grpc.KeepaliveParams(kasp))
+
+	store, cleanUp, err := store.NewTestStoreFromSQL(context.Background(), testFile, config.Datadir)
+	if err != nil {
+		return nil, "", err
+	}
+	t.Cleanup(cleanUp)
+
+	eventStore := &activity.InMemoryEventStore{}
+	if err != nil {
+		return nil, "", err
+	}
+
+	permissionsManager := permissions.NewManager(store)
+	peersManager := peers.NewManager(store, permissionsManager)
+	jobManager := job.NewJobManager(nil, store, peersManager)
+
+	cacheStore, err := nbcache.NewStore(context.Background(), 100*time.Millisecond, 300*time.Millisecond, 100)
+	if err != nil {
+		return nil, "", err
+	}
+
+	ia, _ := validator.NewIntegratedValidator(context.Background(), peersManager, nil, eventStore, cacheStore)
+
+	metrics, err := telemetry.NewDefaultAppMetrics(context.Background())
+	require.NoError(t, err)
+
+	ctrl := gomock.NewController(t)
+	t.Cleanup(ctrl.Finish)
+	settingsMockManager := settings.NewMockManager(ctrl)
+	settingsMockManager.EXPECT().
+		GetSettings(gomock.Any(), gomock.Any(), gomock.Any()).
+		Return(&types.Settings{}, nil).
+		AnyTimes()
+	settingsMockManager.EXPECT().
+		GetExtraSettings(gomock.Any(), gomock.Any()).
+		Return(&types.ExtraSettings{}, nil).
+		AnyTimes()
+
+	groupsManager := groups.NewManagerMock()
+
+	updateManager := update_channel.NewPeersUpdateManager(metrics)
+	requestBuffer := server.NewAccountRequestBuffer(context.Background(), store)
+	networkMapController := controller.NewController(context.Background(), store, metrics, updateManager, requestBuffer, server.MockIntegratedValidator{}, settingsMockManager, "netbird.selfhosted", port_forwarding.NewControllerMock(), manager.NewEphemeralManager(store, peersManager), config)
+	accountManager, err := server.BuildManager(context.Background(), config, store, networkMapController, jobManager, nil, "", eventStore, nil, false, ia, metrics, port_forwarding.NewControllerMock(), settingsMockManager, permissionsManager, false, cacheStore)
+	if err != nil {
+		return nil, "", err
+	}
+
+	secretsManager, err := nbgrpc.NewTimeBasedAuthSecretsManager(updateManager, config.TURNConfig, config.Relay, settingsMockManager, groupsManager)
+	if err != nil {
+		return nil, "", err
+	}
+	mgmtServer, err := nbgrpc.NewServer(config, accountManager, settingsMockManager, jobManager, secretsManager, nil, nil, &server.MockIntegratedValidator{}, networkMapController, nil, nil)
+	if err != nil {
+		return nil, "", err
+	}
+	mgmtProto.RegisterManagementServiceServer(s, mgmtServer)
+	go func() {
+		if err = s.Serve(lis); err != nil {
+			log.Fatalf("failed to serve: %v", err)
+		}
+	}()
+
+	return s, lis.Addr().String(), nil
+}
+
+// getConnectedPeers returns a connection Status or nil if peer connection wasn't found
+func getConnectedPeers(e *Engine) int {
+	e.syncMsgMux.Lock()
+	defer e.syncMsgMux.Unlock()
+	i := 0
+	for _, id := range e.peerStore.PeersPubKey() {
+		conn, _ := e.peerStore.PeerConn(id)
+		if conn.IsConnected() {
+			i++
+		}
+	}
+	return i
+}
+
+func getPeers(e *Engine) int {
+	e.syncMsgMux.Lock()
+	defer e.syncMsgMux.Unlock()
+
+	return len(e.peerStore.PeersPubKey())
+}

client/internal/engine_test.go

@@ -6,37 +6,18 @@ import (
 	"net"
 	"net/netip"
 	"os"
-	"runtime"
 	"strings"
 	"sync"
 	"testing"
 	"time"
 
-	"github.com/golang/mock/gomock"
-	"github.com/google/uuid"
-	log "github.com/sirupsen/logrus"
 	"github.com/stretchr/testify/assert"
 	"github.com/stretchr/testify/require"
-	"go.opentelemetry.io/otel"
 	wgdevice "golang.zx2c4.com/wireguard/device"
 	"golang.zx2c4.com/wireguard/tun/netstack"
 	"golang.zx2c4.com/wireguard/wgctrl/wgtypes"
-	"google.golang.org/grpc"
-	"google.golang.org/grpc/keepalive"
 
 	"github.com/netbirdio/netbird/client/internal/stdnet"
-	"github.com/netbirdio/netbird/management/server/job"
-
-	"github.com/netbirdio/netbird/management/server/integrations/integrated_validator/validator"
-
-	"github.com/netbirdio/netbird/management/internals/controllers/network_map/controller"
-	"github.com/netbirdio/netbird/management/internals/controllers/network_map/update_channel"
-	"github.com/netbirdio/netbird/management/internals/modules/peers"
-	"github.com/netbirdio/netbird/management/internals/modules/peers/ephemeral/manager"
-	nbgrpc "github.com/netbirdio/netbird/management/internals/shared/grpc"
-
-	"github.com/netbirdio/netbird/management/internals/server/config"
-	"github.com/netbirdio/netbird/management/server/groups"
 
 	"github.com/netbirdio/netbird/client/iface"
 	"github.com/netbirdio/netbird/client/iface/configurer"
@@ -50,18 +31,7 @@ import (
 	icemaker "github.com/netbirdio/netbird/client/internal/peer/ice"
 	"github.com/netbirdio/netbird/client/internal/profilemanager"
 	"github.com/netbirdio/netbird/client/internal/routemanager"
-	nbssh "github.com/netbirdio/netbird/client/ssh"
-	"github.com/netbirdio/netbird/client/system"
 	nbdns "github.com/netbirdio/netbird/dns"
-	"github.com/netbirdio/netbird/management/server"
-	"github.com/netbirdio/netbird/management/server/activity"
-	nbcache "github.com/netbirdio/netbird/management/server/cache"
-	"github.com/netbirdio/netbird/management/server/integrations/port_forwarding"
-	"github.com/netbirdio/netbird/management/server/permissions"
-	"github.com/netbirdio/netbird/management/server/settings"
-	"github.com/netbirdio/netbird/management/server/store"
-	"github.com/netbirdio/netbird/management/server/telemetry"
-	"github.com/netbirdio/netbird/management/server/types"
 	"github.com/netbirdio/netbird/monotime"
 	"github.com/netbirdio/netbird/route"
 	mgmt "github.com/netbirdio/netbird/shared/management/client"
@@ -69,25 +39,9 @@ import (
 	"github.com/netbirdio/netbird/shared/netiputil"
 	relayClient "github.com/netbirdio/netbird/shared/relay/client"
 	signal "github.com/netbirdio/netbird/shared/signal/client"
-	"github.com/netbirdio/netbird/shared/signal/proto"
-	signalServer "github.com/netbirdio/netbird/signal/server"
 	"github.com/netbirdio/netbird/util"
 )
 
-var (
-	kaep = keepalive.EnforcementPolicy{
-		MinTime:             15 * time.Second,
-		PermitWithoutStream: true,
-	}
-
-	kasp = keepalive.ServerParameters{
-		MaxConnectionIdle:     15 * time.Second,
-		MaxConnectionAgeGrace: 5 * time.Second,
-		Time:                  5 * time.Second,
-		Timeout:               2 * time.Second,
-	}
-)
-
 type MockWGIface struct {
 	CreateFunc                 func() error
 	CreateOnAndroidFunc        func(routeRange []string, ip string, domains []string) error
@@ -234,129 +188,6 @@ func TestMain(m *testing.M) {
 	os.Exit(code)
 }
 
-func TestEngine_SSH(t *testing.T) {
-	key, err := wgtypes.GeneratePrivateKey()
-	if err != nil {
-		t.Fatal(err)
-		return
-	}
-
-	sshKey, err := nbssh.GeneratePrivateKey(nbssh.ED25519)
-	if err != nil {
-		t.Fatal(err)
-		return
-	}
-
-	ctx, cancel := context.WithCancel(context.Background())
-	defer cancel()
-
-	relayMgr := relayClient.NewManager(ctx, nil, key.PublicKey().String(), iface.DefaultMTU)
-	engine := NewEngine(
-		ctx, cancel,
-		&EngineConfig{
-			WgIfaceName:      "utun101",
-			WgAddr:           wgaddr.MustParseWGAddress("100.64.0.1/24"),
-			WgPrivateKey:     key,
-			WgPort:           33100,
-			ServerSSHAllowed: true,
-			MTU:              iface.DefaultMTU,
-			SSHKey:           sshKey,
-		},
-		EngineServices{
-			SignalClient:   &signal.MockClient{},
-			MgmClient:      &mgmt.MockClient{},
-			RelayManager:   relayMgr,
-			StatusRecorder: peer.NewRecorder("https://mgm"),
-		},
-		MobileDependency{},
-	)
-
-	engine.dnsServer = &dns.MockServer{
-		UpdateDNSServerFunc: func(serial uint64, update nbdns.Config) error { return nil },
-	}
-
-	err = engine.Start(nil, nil)
-	require.NoError(t, err)
-
-	defer func() {
-		err := engine.Stop()
-		if err != nil {
-			return
-		}
-	}()
-
-	peerWithSSH := &mgmtProto.RemotePeerConfig{
-		WgPubKey:   "MNHf3Ma6z6mdLbriAJbqhX7+nM/B71lgw2+91q3LfhU=",
-		AllowedIps: []string{"100.64.0.21/24"},
-		SshConfig: &mgmtProto.SSHConfig{
-			SshPubKey: []byte("ssh-ed25519 AAAAC3NzaC1lZDI1NTE5AAAAIFATYCqaQw/9id1Qkq3n16JYhDhXraI6Pc1fgB8ynEfQ"),
-		},
-	}
-
-	// SSH server is not enabled so SSH config of a remote peer should be ignored
-	networkMap := &mgmtProto.NetworkMap{
-		Serial:             6,
-		PeerConfig:         nil,
-		RemotePeers:        []*mgmtProto.RemotePeerConfig{peerWithSSH},
-		RemotePeersIsEmpty: false,
-	}
-
-	err = engine.updateNetworkMap(networkMap)
-	require.NoError(t, err)
-
-	assert.Nil(t, engine.sshServer)
-
-	// SSH server is enabled, therefore SSH config should be applied
-	networkMap = &mgmtProto.NetworkMap{
-		Serial: 7,
-		PeerConfig: &mgmtProto.PeerConfig{Address: "100.64.0.1/24",
-			SshConfig: &mgmtProto.SSHConfig{
-				SshEnabled: true,
-				JwtConfig: &mgmtProto.JWTConfig{
-					Issuer:       "test-issuer",
-					Audience:     "test-audience",
-					KeysLocation: "test-keys",
-					MaxTokenAge:  3600,
-				},
-			}},
-		RemotePeers:        []*mgmtProto.RemotePeerConfig{peerWithSSH},
-		RemotePeersIsEmpty: false,
-	}
-
-	err = engine.updateNetworkMap(networkMap)
-	require.NoError(t, err)
-
-	time.Sleep(250 * time.Millisecond)
-	assert.NotNil(t, engine.sshServer)
-
-	// now remove peer
-	networkMap = &mgmtProto.NetworkMap{
-		Serial:             8,
-		RemotePeers:        []*mgmtProto.RemotePeerConfig{},
-		RemotePeersIsEmpty: false,
-	}
-
-	err = engine.updateNetworkMap(networkMap)
-	require.NoError(t, err)
-
-	// time.Sleep(250 * time.Millisecond)
-	assert.NotNil(t, engine.sshServer)
-
-	// now disable SSH server
-	networkMap = &mgmtProto.NetworkMap{
-		Serial: 9,
-		PeerConfig: &mgmtProto.PeerConfig{Address: "100.64.0.1/24",
-			SshConfig: &mgmtProto.SSHConfig{SshEnabled: false}},
-		RemotePeers:        []*mgmtProto.RemotePeerConfig{peerWithSSH},
-		RemotePeersIsEmpty: false,
-	}
-
-	err = engine.updateNetworkMap(networkMap)
-	require.NoError(t, err)
-
-	assert.Nil(t, engine.sshServer)
-}
-
 func TestEngine_SSHUpdateLogic(t *testing.T) {
 	// Test that SSH server start/stop logic works based on config
 	engine := &Engine{
@@ -426,7 +257,7 @@ func TestEngine_UpdateNetworkMap(t *testing.T) {
 		return
 	}
 
-	ctx, cancel := context.WithCancel(context.Background())
+	ctx, cancel := context.WithCancel(CtxInitState(context.Background()))
 	defer cancel()
 
 	relayMgr := relayClient.NewManager(ctx, nil, key.PublicKey().String(), iface.DefaultMTU)
@@ -602,7 +433,7 @@ func TestEngine_UpdateNetworkMap(t *testing.T) {
 
 	for _, c := range []testCase{case1, case2, case3, case4, case5, case6} {
 		t.Run(c.name, func(t *testing.T) {
-			err = engine.updateNetworkMap(c.networkMap)
+			_, err = engine.updateNetworkMap(c.networkMap, maxPeersPerSyncPass, true)
 			if err != nil {
 				t.Fatal(err)
 				return
@@ -629,97 +460,47 @@ func TestEngine_UpdateNetworkMap(t *testing.T) {
 			}
 		})
 	}
-}
 
-func TestEngine_Sync(t *testing.T) {
+	// chunked apply: with a per-pass cap smaller than the number of peers, a
-	key, err := wgtypes.GeneratePrivateKey()
+	// single updateNetworkMap applies one batch and reports more==true; the
-	if err != nil {
+	// caller re-runs until convergence. (engine currently holds 0 peers.)
-		t.Fatal(err)
+	t.Run("chunked add converges over multiple passes", func(t *testing.T) {
-		return
+		nm := &mgmtProto.NetworkMap{
-	}
+			Serial:      6,
-
+			RemotePeers: []*mgmtProto.RemotePeerConfig{peer1, peer2, peer3},
-	ctx, cancel := context.WithCancel(context.Background())
-	defer cancel()
-
-	// feed updates to Engine via mocked Management client
-	updates := make(chan *mgmtProto.SyncResponse)
-	defer close(updates)
-	syncFunc := func(ctx context.Context, info *system.Info, msgHandler func(msg *mgmtProto.SyncResponse) error) error {
-		for msg := range updates {
-			err := msgHandler(msg)
-			if err != nil {
-				t.Fatal(err)
-			}
-		}
-		return nil
-	}
-	relayMgr := relayClient.NewManager(ctx, nil, key.PublicKey().String(), iface.DefaultMTU)
-	engine := NewEngine(ctx, cancel, &EngineConfig{
-		WgIfaceName:  "utun103",
-		WgAddr:       wgaddr.MustParseWGAddress("100.64.0.1/24"),
-		WgPrivateKey: key,
-		WgPort:       33100,
-		MTU:          iface.DefaultMTU,
-	}, EngineServices{
-		SignalClient:   &signal.MockClient{},
-		MgmClient:      &mgmt.MockClient{SyncFunc: syncFunc},
-		RelayManager:   relayMgr,
-		StatusRecorder: peer.NewRecorder("https://mgm"),
-	}, MobileDependency{})
-	engine.ctx = ctx
-
-	engine.dnsServer = &dns.MockServer{
-		UpdateDNSServerFunc: func(serial uint64, update nbdns.Config) error { return nil },
-	}
-
-	defer func() {
-		err := engine.Stop()
-		if err != nil {
-			return
-		}
-	}()
-
-	err = engine.Start(nil, nil)
-	if err != nil {
-		t.Fatal(err)
-		return
-	}
-
-	peer1 := &mgmtProto.RemotePeerConfig{
-		WgPubKey:   "RRHf3Ma6z6mdLbriAJbqhX7+nM/B71lgw2+91q3LfhU=",
-		AllowedIps: []string{"100.64.0.10/24"},
-	}
-	peer2 := &mgmtProto.RemotePeerConfig{
-		WgPubKey:   "LLHf3Ma6z6mdLbriAJbqhX9+nM/B71lgw2+91q3LlhU=",
-		AllowedIps: []string{"100.64.0.11/24"},
-	}
-	peer3 := &mgmtProto.RemotePeerConfig{
-		WgPubKey:   "GGHf3Ma6z6mdLbriAJbqhX9+nM/B71lgw2+91q3LlhU=",
-		AllowedIps: []string{"100.64.0.12/24"},
-	}
-	// 1st update with just 1 peer and serial larger than the current serial of the engine => apply update
-	updates <- &mgmtProto.SyncResponse{
-		NetworkMap: &mgmtProto.NetworkMap{
-			Serial:             10,
-			PeerConfig:         nil,
-			RemotePeers:        []*mgmtProto.RemotePeerConfig{peer1, peer2, peer3},
-			RemotePeersIsEmpty: false,
-		},
-	}
-
-	timeout := time.After(time.Second * 2)
-	for {
-		select {
-		case <-timeout:
-			t.Fatalf("timeout while waiting for test to finish")
-			return
-		default:
 		}
 
-		if getPeers(engine) == 3 && engine.networkSerial == 10 {
+		more, err := engine.updateNetworkMap(nm, 1, true)
-			break
+		require.NoError(t, err)
+		require.True(t, more, "pass 1 should signal more")
+		require.Len(t, engine.peerStore.PeersPubKey(), 1)
+
+		more, err = engine.updateNetworkMap(nm, 1, false)
+		require.NoError(t, err)
+		require.True(t, more, "pass 2 should signal more")
+		require.Len(t, engine.peerStore.PeersPubKey(), 2)
+
+		more, err = engine.updateNetworkMap(nm, 1, false)
+		require.NoError(t, err)
+		require.False(t, more, "pass 3 should converge")
+		require.Len(t, engine.peerStore.PeersPubKey(), 3)
+	})
+
+	t.Run("chunked remove converges over multiple passes", func(t *testing.T) {
+		nm := &mgmtProto.NetworkMap{
+			Serial:      7,
+			RemotePeers: []*mgmtProto.RemotePeerConfig{peer1}, // remove peer2, peer3
 		}
-	}
+
+		more, err := engine.updateNetworkMap(nm, 1, true)
+		require.NoError(t, err)
+		require.True(t, more, "pass 1 should signal more (2 to remove, cap 1)")
+
+		more, err = engine.updateNetworkMap(nm, 1, false)
+		require.NoError(t, err)
+		require.False(t, more, "pass 2 should converge")
+		require.Len(t, engine.peerStore.PeersPubKey(), 1)
+	})
 }
 
 func TestEngine_UpdateNetworkMapWithRoutes(t *testing.T) {
@@ -817,7 +598,7 @@ func TestEngine_UpdateNetworkMapWithRoutes(t *testing.T) {
 				return
 			}
 
-			ctx, cancel := context.WithCancel(context.Background())
+			ctx, cancel := context.WithCancel(CtxInitState(context.Background()))
 			defer cancel()
 
 			wgIfaceName := fmt.Sprintf("utun%d", 104+n)
@@ -890,7 +671,7 @@ func TestEngine_UpdateNetworkMapWithRoutes(t *testing.T) {
 				}
 			}()
 
-			err = engine.updateNetworkMap(testCase.networkMap)
+			_, err = engine.updateNetworkMap(testCase.networkMap, maxPeersPerSyncPass, true)
 			assert.NoError(t, err, "shouldn't return error")
 			assert.Equal(t, testCase.expectedSerial, input.inputSerial, "serial should match")
 			assert.Len(t, input.clientRoutes, testCase.expectedLen, "clientRoutes len should match")
@@ -1024,7 +805,7 @@ func TestEngine_UpdateNetworkMapWithDNSUpdate(t *testing.T) {
 				return
 			}
 
-			ctx, cancel := context.WithCancel(context.Background())
+			ctx, cancel := context.WithCancel(CtxInitState(context.Background()))
 			defer cancel()
 
 			wgIfaceName := fmt.Sprintf("utun%d", 104+n)
@@ -1094,7 +875,7 @@ func TestEngine_UpdateNetworkMapWithDNSUpdate(t *testing.T) {
 				}
 			}()
 
-			err = engine.updateNetworkMap(testCase.networkMap)
+			_, err = engine.updateNetworkMap(testCase.networkMap, maxPeersPerSyncPass, true)
 			assert.NoError(t, err, "shouldn't return error")
 			assert.Equal(t, testCase.expectedSerial, input.inputSerial, "serial should match")
 			assert.Len(t, input.inputNSGroups, testCase.expectedZonesLen, "zones len should match")
@@ -1105,104 +886,6 @@ func TestEngine_UpdateNetworkMapWithDNSUpdate(t *testing.T) {
 	}
 }
 
-func TestEngine_MultiplePeers(t *testing.T) {
-	// log.SetLevel(log.DebugLevel)
-
-	ctx, cancel := context.WithCancel(CtxInitState(context.Background()))
-	defer cancel()
-
-	sigServer, signalAddr, err := startSignal(t)
-	if err != nil {
-		t.Fatal(err)
-		return
-	}
-	defer sigServer.Stop()
-	mgmtServer, mgmtAddr, err := startManagement(t, t.TempDir(), "../testdata/store.sql")
-	if err != nil {
-		t.Fatal(err)
-		return
-	}
-	defer mgmtServer.GracefulStop()
-
-	setupKey := "A2C8E62B-38F5-4553-B31E-DD66C696CEBB"
-
-	mu := sync.Mutex{}
-	engines := []*Engine{}
-	numPeers := 10
-	wg := sync.WaitGroup{}
-	wg.Add(numPeers)
-	// create and start peers
-	for i := 0; i < numPeers; i++ {
-		j := i
-		go func() {
-			engine, err := createEngine(ctx, cancel, setupKey, j, mgmtAddr, signalAddr)
-			if err != nil {
-				wg.Done()
-				t.Errorf("unable to create the engine for peer %d with error %v", j, err)
-				return
-			}
-			engine.dnsServer = &dns.MockServer{}
-			mu.Lock()
-			defer mu.Unlock()
-			guid := fmt.Sprintf("{%s}", uuid.New().String())
-			device.CustomWindowsGUIDString = strings.ToLower(guid)
-			err = engine.Start(nil, nil)
-			if err != nil {
-				t.Errorf("unable to start engine for peer %d with error %v", j, err)
-				wg.Done()
-				return
-			}
-			engines = append(engines, engine)
-			wg.Done()
-		}()
-	}
-
-	// wait until all have been created and started
-	wg.Wait()
-	if len(engines) != numPeers {
-		t.Fatal("not all peers was started")
-	}
-	// check whether all the peer have expected peers connected
-
-	expectedConnected := numPeers * (numPeers - 1)
-
-	// adjust according to timeouts
-	timeout := 50 * time.Second
-	timeoutChan := time.After(timeout)
-	ticker := time.NewTicker(time.Second)
-	defer ticker.Stop()
-loop:
-	for {
-		select {
-		case <-timeoutChan:
-			t.Fatalf("waiting for expected connections timeout after %s", timeout.String())
-			break loop
-		case <-ticker.C:
-			totalConnected := 0
-			for _, engine := range engines {
-				totalConnected += getConnectedPeers(engine)
-			}
-			if totalConnected == expectedConnected {
-				log.Infof("total connected=%d", totalConnected)
-				break loop
-			}
-			log.Infof("total connected=%d", totalConnected)
-		}
-	}
-	// cleanup test
-	for n, peerEngine := range engines {
-		t.Logf("stopping peer with interface %s from multipeer test, loopIndex %d", peerEngine.wgInterface.Name(), n)
-		errStop := peerEngine.mgmClient.Close()
-		if errStop != nil {
-			log.Infoln("got error trying to close management clients from engine: ", errStop)
-		}
-		errStop = peerEngine.Stop()
-		if errStop != nil {
-			log.Infoln("got error trying to close testing peers engine: ", errStop)
-		}
-	}
-}
-
 func Test_ParseNATExternalIPMappings(t *testing.T) {
 	ifaceList, err := net.Interfaces()
 	if err != nil {
@@ -1526,187 +1209,6 @@ func TestCompareNetIPLists(t *testing.T) {
 	}
 }
 
-func createEngine(ctx context.Context, cancel context.CancelFunc, setupKey string, i int, mgmtAddr string, signalAddr string) (*Engine, error) {
-	key, err := wgtypes.GeneratePrivateKey()
-	if err != nil {
-		return nil, err
-	}
-	mgmtClient, err := mgmt.NewClient(ctx, mgmtAddr, key, false)
-	if err != nil {
-		return nil, err
-	}
-	signalClient, err := signal.NewClient(ctx, signalAddr, key, false)
-	if err != nil {
-		return nil, err
-	}
-
-	info := system.GetInfo(ctx)
-	resp, err := mgmtClient.Register(setupKey, "", info, nil, nil)
-	if err != nil {
-		return nil, err
-	}
-
-	var ifaceName string
-	if runtime.GOOS == "darwin" {
-		ifaceName = fmt.Sprintf("utun1%d", i)
-	} else {
-		ifaceName = fmt.Sprintf("wt%d", i)
-	}
-
-	wgPort := 33100 + i
-	conf := &EngineConfig{
-		WgIfaceName:  ifaceName,
-		WgAddr:       wgaddr.MustParseWGAddress(resp.PeerConfig.Address),
-		WgPrivateKey: key,
-		WgPort:       wgPort,
-		MTU:          iface.DefaultMTU,
-	}
-
-	relayMgr := relayClient.NewManager(ctx, nil, key.PublicKey().String(), iface.DefaultMTU)
-	e, err := NewEngine(ctx, cancel, conf, EngineServices{
-		SignalClient:   signalClient,
-		MgmClient:      mgmtClient,
-		RelayManager:   relayMgr,
-		StatusRecorder: peer.NewRecorder("https://mgm"),
-	}, MobileDependency{}), nil
-	e.ctx = ctx
-	return e, err
-}
-
-func startSignal(t *testing.T) (*grpc.Server, string, error) {
-	t.Helper()
-
-	s := grpc.NewServer(grpc.KeepaliveEnforcementPolicy(kaep), grpc.KeepaliveParams(kasp))
-
-	lis, err := net.Listen("tcp", "localhost:0")
-	if err != nil {
-		log.Fatalf("failed to listen: %v", err)
-	}
-
-	srv, err := signalServer.NewServer(context.Background(), otel.Meter(""))
-	require.NoError(t, err)
-	proto.RegisterSignalExchangeServer(s, srv)
-
-	go func() {
-		if err = s.Serve(lis); err != nil {
-			log.Fatalf("failed to serve: %v", err)
-		}
-	}()
-
-	return s, lis.Addr().String(), nil
-}
-
-func startManagement(t *testing.T, dataDir, testFile string) (*grpc.Server, string, error) {
-	t.Helper()
-
-	config := &config.Config{
-		Stuns:      []*config.Host{},
-		TURNConfig: &config.TURNConfig{},
-		Relay: &config.Relay{
-			Addresses:      []string{"127.0.0.1:1234"},
-			CredentialsTTL: util.Duration{Duration: time.Hour},
-			Secret:         "222222222222222222",
-		},
-		Signal: &config.Host{
-			Proto: "http",
-			URI:   "localhost:10000",
-		},
-		Datadir:    dataDir,
-		HttpConfig: nil,
-	}
-
-	lis, err := net.Listen("tcp", "localhost:0")
-	if err != nil {
-		return nil, "", err
-	}
-	s := grpc.NewServer(grpc.KeepaliveEnforcementPolicy(kaep), grpc.KeepaliveParams(kasp))
-
-	store, cleanUp, err := store.NewTestStoreFromSQL(context.Background(), testFile, config.Datadir)
-	if err != nil {
-		return nil, "", err
-	}
-	t.Cleanup(cleanUp)
-
-	eventStore := &activity.InMemoryEventStore{}
-	if err != nil {
-		return nil, "", err
-	}
-
-	permissionsManager := permissions.NewManager(store)
-	peersManager := peers.NewManager(store, permissionsManager)
-	jobManager := job.NewJobManager(nil, store, peersManager)
-
-	cacheStore, err := nbcache.NewStore(context.Background(), 100*time.Millisecond, 300*time.Millisecond, 100)
-	if err != nil {
-		return nil, "", err
-	}
-
-	ia, _ := validator.NewIntegratedValidator(context.Background(), peersManager, nil, eventStore, cacheStore)
-
-	metrics, err := telemetry.NewDefaultAppMetrics(context.Background())
-	require.NoError(t, err)
-
-	ctrl := gomock.NewController(t)
-	t.Cleanup(ctrl.Finish)
-	settingsMockManager := settings.NewMockManager(ctrl)
-	settingsMockManager.EXPECT().
-		GetSettings(gomock.Any(), gomock.Any(), gomock.Any()).
-		Return(&types.Settings{}, nil).
-		AnyTimes()
-	settingsMockManager.EXPECT().
-		GetExtraSettings(gomock.Any(), gomock.Any()).
-		Return(&types.ExtraSettings{}, nil).
-		AnyTimes()
-
-	groupsManager := groups.NewManagerMock()
-
-	updateManager := update_channel.NewPeersUpdateManager(metrics)
-	requestBuffer := server.NewAccountRequestBuffer(context.Background(), store)
-	networkMapController := controller.NewController(context.Background(), store, metrics, updateManager, requestBuffer, server.MockIntegratedValidator{}, settingsMockManager, "netbird.selfhosted", port_forwarding.NewControllerMock(), manager.NewEphemeralManager(store, peersManager), config)
-	accountManager, err := server.BuildManager(context.Background(), config, store, networkMapController, jobManager, nil, "", eventStore, nil, false, ia, metrics, port_forwarding.NewControllerMock(), settingsMockManager, permissionsManager, false, cacheStore)
-	if err != nil {
-		return nil, "", err
-	}
-
-	secretsManager, err := nbgrpc.NewTimeBasedAuthSecretsManager(updateManager, config.TURNConfig, config.Relay, settingsMockManager, groupsManager)
-	if err != nil {
-		return nil, "", err
-	}
-	mgmtServer, err := nbgrpc.NewServer(config, accountManager, settingsMockManager, jobManager, secretsManager, nil, nil, &server.MockIntegratedValidator{}, networkMapController, nil, nil)
-	if err != nil {
-		return nil, "", err
-	}
-	mgmtProto.RegisterManagementServiceServer(s, mgmtServer)
-	go func() {
-		if err = s.Serve(lis); err != nil {
-			log.Fatalf("failed to serve: %v", err)
-		}
-	}()
-
-	return s, lis.Addr().String(), nil
-}
-
-// getConnectedPeers returns a connection Status or nil if peer connection wasn't found
-func getConnectedPeers(e *Engine) int {
-	e.syncMsgMux.Lock()
-	defer e.syncMsgMux.Unlock()
-	i := 0
-	for _, id := range e.peerStore.PeersPubKey() {
-		conn, _ := e.peerStore.PeerConn(id)
-		if conn.IsConnected() {
-			i++
-		}
-	}
-	return i
-}
-
-func getPeers(e *Engine) int {
-	e.syncMsgMux.Lock()
-	defer e.syncMsgMux.Unlock()
-
-	return len(e.peerStore.PeersPubKey())
-}
-
 func mustEncodePrefix(t *testing.T, p netip.Prefix) []byte {
 	t.Helper()
 	b, err := netiputil.EncodePrefix(p)

client/internal/lazyconn/activity/listener_bind.go

@@ -119,10 +119,6 @@ func (d *BindListener) ReadPackets() {
 	}
 
 	d.peerCfg.Log.Debugf("removing lazy endpoint for peer %s", d.peerCfg.PublicKey)
-	if err := d.wgIface.RemovePeer(d.peerCfg.PublicKey); err != nil {
-		d.peerCfg.Log.Errorf("failed to remove endpoint: %s", err)
-	}
-
 	_ = d.lazyConn.Close()
 	d.bind.RemoveEndpoint(d.fakeIP)
 	d.done.Done()

client/internal/mapsync.go

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

client/internal/mapsync_test.go

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

client/internal/peer/conn_status.go

@@ -26,7 +26,6 @@ type connStatusInputs struct {
 	iceInProgress       bool // a negotiation is currently in flight
 }
 
-
 // ConnStatus describe the status of a peer's connection
 type ConnStatus int32
 

client/internal/peer/handshaker.go

@@ -195,14 +195,14 @@ func (h *Handshaker) sendOffer() error {
 	}
 
 	offer := h.buildOfferAnswer()
-	h.log.Infof("sending offer with serial: %s", offer.SessionIDString())
+	h.log.Debugf("sending offer with serial: %s", offer.SessionIDString())
 
 	return h.signaler.SignalOffer(offer, h.config.Key)
 }
 
 func (h *Handshaker) sendAnswer() error {
 	answer := h.buildOfferAnswer()
-	h.log.Infof("sending answer with serial: %s", answer.SessionIDString())
+	h.log.Debugf("sending answer with serial: %s", answer.SessionIDString())
 
 	return h.signaler.SignalAnswer(answer, h.config.Key)
 }

client/internal/peer/status.go

@@ -192,7 +192,9 @@ func (s *StatusChangeSubscription) Events() chan map[string]RouterState {
 // Pure read methods take RLock; anything that mutates state takes Lock.
 type Status struct {
 	mux                   sync.RWMutex
+	muxRelays             sync.RWMutex
 	peers                 map[string]State
+	ipToKey               map[string]string
 	changeNotify          map[string]map[string]*StatusChangeSubscription // map[peerID]map[subscriptionID]*StatusChangeSubscription
 	signalState           bool
 	signalError           error
@@ -231,6 +233,7 @@ type Status struct {
 func NewRecorder(mgmAddress string) *Status {
 	return &Status{
 		peers:                 make(map[string]State),
+		ipToKey:               make(map[string]string),
 		changeNotify:          make(map[string]map[string]*StatusChangeSubscription),
 		eventStreams:          make(map[string]chan *proto.SystemEvent),
 		eventQueue:            NewEventQueue(eventQueueSize),
@@ -242,8 +245,8 @@ func NewRecorder(mgmAddress string) *Status {
 }
 
 func (d *Status) SetRelayMgr(manager *relayClient.Manager) {
-	d.mux.Lock()
+	d.muxRelays.Lock()
-	defer d.mux.Unlock()
+	defer d.muxRelays.Unlock()
 	d.relayMgr = manager
 }
 
@@ -282,6 +285,12 @@ func (d *Status) AddPeer(peerPubKey string, fqdn string, ip string, ipv6 string)
 		Mux:        new(sync.RWMutex),
 	}
 	d.peerListChangedForNotification = true
+	if ipv6 != "" {
+		d.ipToKey[ipv6] = peerPubKey
+	}
+	if ip != "" {
+		d.ipToKey[ip] = peerPubKey
+	}
 	return nil
 }
 
@@ -311,28 +320,22 @@ func (d *Status) PeerByIP(ip string) (string, bool) {
 
 // PeerStateByIP returns the full peer State for the given tunnel IP.
 // Matches against either the IPv4 (State.IP) or IPv6 (State.IPv6) tunnel
-// address so dual-stack peers are reachable on either family. Searches
+// address so dual-stack peers are reachable on either family. Only
-// both d.peers and d.offlinePeers — peers that have been moved into
+// active peers are matched; peers moved into the offline slice by
-// the offline slice by ReplaceOfflinePeers are still part of the
+// ReplaceOfflinePeers are intentionally treated as unknown.
-// account's roster and callers (DNS filter, embed.Client.IdentityForIP)
-// need to recognise them rather than treating them as unknown. Returns
-// the zero State and false when no peer matches or the input is empty.
 func (d *Status) PeerStateByIP(ip string) (State, bool) {
 	if ip == "" {
 		return State{}, false
 	}
 	d.mux.RLock()
 	defer d.mux.RUnlock()
-
+	key, ok := d.ipToKey[ip]
-	for _, state := range d.peers {
+	if !ok {
-		if (state.IP != "" && state.IP == ip) || (state.IPv6 != "" && state.IPv6 == ip) {
+		return State{}, false
-			return state, true
-		}
 	}
-	for _, state := range d.offlinePeers {
+	state, ok := d.peers[key]
-		if (state.IP != "" && state.IP == ip) || (state.IPv6 != "" && state.IPv6 == ip) {
+	if ok {
-			return state, true
+		return state, true
-		}
 	}
 	return State{}, false
 }
@@ -342,12 +345,18 @@ func (d *Status) RemovePeer(peerPubKey string) error {
 	d.mux.Lock()
 	defer d.mux.Unlock()
 
-	_, ok := d.peers[peerPubKey]
+	p, ok := d.peers[peerPubKey]
 	if !ok {
 		return errors.New("no peer with to remove")
 	}
 
 	delete(d.peers, peerPubKey)
+	if mappedKey, exists := d.ipToKey[p.IP]; exists && mappedKey == peerPubKey {
+		delete(d.ipToKey, p.IP)
+	}
+	if mappedKey, exists := d.ipToKey[p.IPv6]; exists && mappedKey == peerPubKey {
+		delete(d.ipToKey, p.IPv6)
+	}
 	d.peerListChangedForNotification = true
 	return nil
 }
@@ -898,8 +907,8 @@ func (d *Status) MarkSignalConnected() {
 }
 
 func (d *Status) UpdateRelayStates(relayResults []relay.ProbeResult) {
-	d.mux.Lock()
+	d.muxRelays.Lock()
-	defer d.mux.Unlock()
+	defer d.muxRelays.Unlock()
 	d.relayStates = relayResults
 }
 
@@ -1010,21 +1019,26 @@ func (d *Status) GetSignalState() SignalState {
 
 // GetRelayStates returns the stun/turn/permanent relay states
 func (d *Status) GetRelayStates() []relay.ProbeResult {
-	d.mux.RLock()
+	d.muxRelays.RLock()
-	defer d.mux.RUnlock()
 	if d.relayMgr == nil {
-		return d.relayStates
+		defer d.muxRelays.RUnlock()
+		return slices.Clone(d.relayStates)
 	}
 
-	// extend the list of stun, turn servers with relay address
+	relayMgr := d.relayMgr
+	// extend the list of stun, turn servers with the relay server connections
 	relayStates := slices.Clone(d.relayStates)
+	d.muxRelays.RUnlock()
 
-	// if the server connection is not established then we will use the general address
+	states := relayMgr.RelayStates()
-	// in case of connection we will use the instance specific address
+	if len(states) == 0 {
-	instanceAddr, _, err := d.relayMgr.RelayInstanceAddress()
+		// no relay connection tracked yet; surface configured servers as
-	if err != nil {
+		// unavailable with the real reconnect error when known
-		// TODO add their status
+		err := relayClient.ErrRelayClientNotConnected
-		for _, r := range d.relayMgr.ServerURLs() {
+		if connErr := relayMgr.RelayConnectError(); connErr != nil {
+			err = connErr
+		}
+		for _, r := range relayMgr.ServerURLs() {
 			relayStates = append(relayStates, relay.ProbeResult{
 				URI: r,
 				Err: err,
@@ -1033,10 +1047,14 @@ func (d *Status) GetRelayStates() []relay.ProbeResult {
 		return relayStates
 	}
 
-	relayState := relay.ProbeResult{
+	for _, rs := range states {
-		URI: instanceAddr,
+		relayStates = append(relayStates, relay.ProbeResult{
+			URI:       rs.URL,
+			Err:       rs.Err,
+			Transport: rs.Transport,
+		})
 	}
-	return append(relayStates, relayState)
+	return relayStates
 }
 
 func (d *Status) ForwardingRules() []firewall.ForwardRule {
@@ -1397,6 +1415,7 @@ func (fs FullStatus) ToProto() *proto.FullStatus {
 		pbRelayState := &proto.RelayState{
 			URI:       relayState.URI,
 			Available: relayState.Err == nil,
+			Transport: relayState.Transport,
 		}
 		if err := relayState.Err; err != nil {
 			pbRelayState.Error = err.Error()

client/internal/peer/status_test.go

@@ -90,12 +90,11 @@ func TestStatus_PeerStateByIP_MatchesIPv6(t *testing.T) {
 	req.Equal("pk-1", state.PubKey, "matching state must carry the right pub key")
 }
 
-// TestStatus_PeerStateByIP_MatchesOfflinePeers covers peers that have
+// TestStatus_PeerStateByIP_IgnoresOfflinePeers documents that peers
-// been moved into the offline slice via ReplaceOfflinePeers. Callers
+// moved into the offline slice via ReplaceOfflinePeers are intentionally
-// (DNS filter, embed.Client.IdentityForIP) need to treat them as known
+// not resolvable by IP: only active peers can carry traffic, so callers
-// rather than unknown — otherwise authentication / DNS filtering treats
+// (DNS filter, embed.Client.IdentityForIP) treat them as unknown.
-// known-but-offline peers as foreign IPs.
+func TestStatus_PeerStateByIP_IgnoresOfflinePeers(t *testing.T) {
-func TestStatus_PeerStateByIP_MatchesOfflinePeers(t *testing.T) {
 	status := NewRecorder("https://mgm")
 	req := require.New(t)
 
@@ -103,13 +102,31 @@ func TestStatus_PeerStateByIP_MatchesOfflinePeers(t *testing.T) {
 		{PubKey: "pk-offline", FQDN: "offline.netbird", IP: "100.64.0.20", IPv6: "fd00::20"},
 	})
 
-	state, ok := status.PeerStateByIP("100.64.0.20")
+	_, ok := status.PeerStateByIP("100.64.0.20")
-	req.True(ok, "offline peer must resolve by IPv4 tunnel address")
+	req.False(ok, "offline peer must not resolve by IPv4 tunnel address")
-	req.Equal("pk-offline", state.PubKey, "matching state must carry the offline peer's pub key")
 
-	state, ok = status.PeerStateByIP("fd00::20")
+	_, ok = status.PeerStateByIP("fd00::20")
-	req.True(ok, "offline peer must resolve by IPv6 tunnel address")
+	req.False(ok, "offline peer must not resolve by IPv6 tunnel address")
-	req.Equal("pk-offline", state.PubKey, "IPv6 match must carry the offline peer's pub key")
+}
+
+// TestStatus_PeerStateByIP_RemovedPeer verifies RemovePeer drops the
+// IP index entries for both address families.
+func TestStatus_PeerStateByIP_RemovedPeer(t *testing.T) {
+	status := NewRecorder("https://mgm")
+	req := require.New(t)
+
+	req.NoError(status.AddPeer("pk-1", "peer-1.netbird", "100.64.0.10", "fd00::1"))
+
+	_, ok := status.PeerStateByIP("100.64.0.10")
+	req.True(ok, "active peer must resolve before removal")
+
+	req.NoError(status.RemovePeer("pk-1"))
+
+	_, ok = status.PeerStateByIP("100.64.0.10")
+	req.False(ok, "removed peer must not resolve by IPv4 tunnel address")
+
+	_, ok = status.PeerStateByIP("fd00::1")
+	req.False(ok, "removed peer must not resolve by IPv6 tunnel address")
 }
 
 func TestStatus_UpdatePeerFQDN(t *testing.T) {

client/internal/peer/worker_ice.go

@@ -4,7 +4,6 @@ import (
 	"context"
 	"fmt"
 	"net"
-	"net/netip"
 	"strconv"
 	"sync"
 	"time"
@@ -165,10 +164,6 @@ func (w *WorkerICE) OnRemoteCandidate(candidate ice.Candidate, haRoutes route.HA
 		return
 	}
 
-	if candidateViaRoutes(candidate, haRoutes) {
-		return
-	}
-
 	if err := w.agent.AddRemoteCandidate(candidate); err != nil {
 		w.log.Errorf("error while handling remote candidate")
 		return
@@ -589,34 +584,6 @@ func extraSrflxCandidate(candidate ice.Candidate) (*ice.CandidateServerReflexive
 	return ec, nil
 }
 
-func candidateViaRoutes(candidate ice.Candidate, clientRoutes route.HAMap) bool {
-	addr, err := netip.ParseAddr(candidate.Address())
-	if err != nil {
-		log.Errorf("Failed to parse IP address %s: %v", candidate.Address(), err)
-		return false
-	}
-
-	var routePrefixes []netip.Prefix
-	for _, routes := range clientRoutes {
-		if len(routes) > 0 && routes[0] != nil {
-			routePrefixes = append(routePrefixes, routes[0].Network)
-		}
-	}
-
-	for _, prefix := range routePrefixes {
-		// default route is handled by route exclusion / ip rules
-		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
-}
-
 func isRelayCandidate(candidate ice.Candidate) bool {
 	return candidate.Type() == ice.CandidateTypeRelay
 }

client/internal/profilemanager/config.go

@@ -22,6 +22,7 @@ import (
 
 	"github.com/netbirdio/netbird/client/iface"
 	"github.com/netbirdio/netbird/client/internal/routemanager/dynamic"
+	"github.com/netbirdio/netbird/client/mdm"
 	"github.com/netbirdio/netbird/client/ssh"
 	mgm "github.com/netbirdio/netbird/shared/management/client"
 	"github.com/netbirdio/netbird/shared/management/domain"
@@ -57,6 +58,10 @@ var DefaultInterfaceBlacklist = []string{
 	"Tailscale", "tailscale", "docker", "veth", "br-", "lo",
 }
 
+// loadMDMPolicy is the package-level indirection used by apply() to read the
+// active MDM policy. Tests override this to inject a fake policy.
+var loadMDMPolicy = mdm.LoadPolicy
+
 // ConfigInput carries configuration changes to the client
 type ConfigInput struct {
 	ManagementURL                 string
@@ -103,6 +108,10 @@ type ConfigInput struct {
 
 // Config Configuration type
 type Config struct {
+	// Name is the human-readable profile name shown in CLI/UI listings.
+	// It is independent of the profile's on-disk filename (which is the ID).
+	Name string
+
 	// Wireguard private key of local peer
 	PrivateKey                    string
 	PreSharedKey                  string
@@ -174,6 +183,23 @@ type Config struct {
 	LazyConnectionEnabled bool
 
 	MTU uint16
+
+	// policy is the MDM policy that produced the currently-set values for
+	// any MDM-enforced fields. Set by applyMDMPolicy at the tail of apply()
+	// and reset on every apply() invocation. Never persisted to disk.
+	// Callers query enforcement state via Policy() and the mdm.Policy API
+	// (HasKey, ManagedKeys, IsEmpty).
+	policy *mdm.Policy `json:"-"`
+}
+
+// Policy returns the MDM policy applied to this Config. Returns a non-nil
+// empty Policy when MDM enforcement is inactive; callers can always invoke
+// HasKey / ManagedKeys / IsEmpty without a nil check.
+func (config *Config) Policy() *mdm.Policy {
+	if config == nil || config.policy == nil {
+		return mdm.NewPolicy(nil)
+	}
+	return config.policy
 }
 
 var ConfigDirOverride string
@@ -248,6 +274,16 @@ func createNewConfig(input ConfigInput) (*Config, error) {
 }
 
 func (config *Config) apply(input ConfigInput) (updated bool, err error) {
+	if config.Name != "" {
+		sanitized, err := sanitizeDisplayName(config.Name)
+		if err != nil {
+			return false, fmt.Errorf("invalid profile name: %w", err)
+		}
+		if sanitized != config.Name {
+			config.Name = sanitized
+			updated = true
+		}
+	}
 	if config.ManagementURL == nil {
 		log.Infof("using default Management URL %s", DefaultManagementURL)
 		config.ManagementURL, err = parseURL("Management URL", DefaultManagementURL)
@@ -397,7 +433,7 @@ func (config *Config) apply(input ConfigInput) (updated bool, err error) {
 		updated = true
 	}
 
-	if input.ServerSSHAllowed != nil && *input.ServerSSHAllowed != *config.ServerSSHAllowed {
+	if input.ServerSSHAllowed != nil && (config.ServerSSHAllowed == nil || *input.ServerSSHAllowed != *config.ServerSSHAllowed) {
 		if *input.ServerSSHAllowed {
 			log.Infof("enabling SSH server")
 		} else {
@@ -612,10 +648,93 @@ func (config *Config) apply(input ConfigInput) (updated bool, err error) {
 		updated = true
 	}
 
+	// MDM is the last override layer: any key present in the policy
+	// supersedes defaults, on-disk config, env vars and CLI input.
+	config.applyMDMPolicy(loadMDMPolicy())
+
 	return updated, nil
 }
 
-// parseURL parses and validates a service URL
+// applyMDMPolicy overlays MDM-supplied values on top of the resolved Config.
+// The provided Policy is also stored on the Config so callers can later query
+// which fields are enforced. Invalid values (e.g. malformed URLs) are logged
+// and skipped to avoid bricking the client; the field keeps its previous
+// resolved value but is still marked as managed (Policy.HasKey returns true
+// for the key, so per-field rejection of user writes still applies).
+func (config *Config) applyMDMPolicy(policy *mdm.Policy) {
+	config.policy = policy
+	if policy.IsEmpty() {
+		return
+	}
+
+	// Helper: log the application of a single MDM-managed key. Values for
+	// keys in mdm.SecretKeys are redacted.
+	logApplied := func(key string, displayValue any) {
+		if _, secret := mdm.SecretKeys[key]; secret {
+			log.Infof("MDM override %s = ********** (secret)", key)
+			return
+		}
+		log.Infof("MDM override %s = %v", key, displayValue)
+	}
+
+	if v, ok := policy.GetString(mdm.KeyManagementURL); ok {
+		if u, err := parseURL("Management URL", v); err != nil {
+			log.Warnf("MDM management URL %q invalid: %v; keeping previous value", v, err)
+		} else {
+			config.ManagementURL = u
+			logApplied(mdm.KeyManagementURL, u.String())
+		}
+	}
+
+	if v, ok := policy.GetString(mdm.KeyPreSharedKey); ok {
+		// Defensive: refuse the redaction mask in case it round-tripped
+		// through a manifest by mistake.
+		if !isPreSharedKeyHidden(&v) {
+			config.PreSharedKey = v
+			logApplied(mdm.KeyPreSharedKey, "")
+		}
+	}
+
+	// applyBool collapses the per-key "read + set + log" boilerplate
+	// for every plain bool MDM key into a single helper. Keeps the
+	// outer function's cognitive complexity below SonarCube's
+	// threshold; functional behaviour is identical to the inlined
+	// branches it replaces.
+	applyBool := func(key string, setter func(bool)) {
+		v, ok := policy.GetBool(key)
+		if !ok {
+			return
+		}
+		setter(v)
+		logApplied(key, v)
+	}
+
+	applyBool(mdm.KeyAllowServerSSH, func(v bool) { bv := v; config.ServerSSHAllowed = &bv })
+	applyBool(mdm.KeyDisableClientRoutes, func(v bool) { config.DisableClientRoutes = v })
+	applyBool(mdm.KeyDisableServerRoutes, func(v bool) { config.DisableServerRoutes = v })
+	applyBool(mdm.KeyBlockInbound, func(v bool) { config.BlockInbound = v })
+	applyBool(mdm.KeyDisableAutoConnect, func(v bool) { config.DisableAutoConnect = v })
+	applyBool(mdm.KeyRosenpassEnabled, func(v bool) { config.RosenpassEnabled = v })
+	applyBool(mdm.KeyRosenpassPermissive, func(v bool) { config.RosenpassPermissive = v })
+
+	if v, ok := policy.GetInt(mdm.KeyWireguardPort); ok {
+		// REG_DWORD is 32-bit; UDP port range is 1-65535. Clamp at the
+		// upper bound and reject obviously-invalid values to avoid the
+		// engine binding to an unusable port if the admin pushes garbage.
+		if v >= 1 && v <= 65535 {
+			config.WgPort = int(v)
+			logApplied(mdm.KeyWireguardPort, v)
+		} else {
+			log.Warnf("MDM wireguard port %d out of range [1,65535]; keeping previous value", v)
+		}
+	}
+}
+
+// parseURL parses and validates the URL for the named service. The URL
+// must use the http or https scheme; if no port is present, ":443" is
+// appended for https or ":80" for http. The serviceName parameter is
+// used to contextualise error messages. On success returns the parsed
+// *url.URL; on failure returns a non-nil error.
 func parseURL(serviceName, serviceURL string) (*url.URL, error) {
 	parsedMgmtURL, err := url.ParseRequestURI(serviceURL)
 	if err != nil {

client/internal/profilemanager/config_mdm_test.go

@@ -0,0 +1,152 @@
+package profilemanager
+
+import (
+	"path/filepath"
+	"testing"
+
+	"github.com/stretchr/testify/assert"
+	"github.com/stretchr/testify/require"
+
+	"github.com/netbirdio/netbird/client/mdm"
+)
+
+// withMDMPolicy temporarily overrides the package-level loadMDMPolicy hook so
+// apply() observes the supplied Policy. The original loader is restored at
+// test cleanup.
+func withMDMPolicy(t *testing.T, policy *mdm.Policy) {
+	t.Helper()
+	prev := loadMDMPolicy
+	loadMDMPolicy = func() *mdm.Policy { return policy }
+	t.Cleanup(func() { loadMDMPolicy = prev })
+}
+
+func TestApply_MDMEmpty_NoEnforcement(t *testing.T) {
+	withMDMPolicy(t, mdm.NewPolicy(nil))
+
+	cfg, err := UpdateOrCreateConfig(ConfigInput{
+		ConfigPath: filepath.Join(t.TempDir(), "config.json"),
+	})
+	require.NoError(t, err)
+	require.NotNil(t, cfg)
+
+	assert.True(t, cfg.Policy().IsEmpty(), "no MDM source ⇒ empty Policy")
+	assert.False(t, cfg.Policy().HasKey(mdm.KeyManagementURL))
+	assert.Empty(t, cfg.Policy().ManagedKeys())
+
+	// Default management URL still resolves.
+	assert.Equal(t, DefaultManagementURL, cfg.ManagementURL.String())
+}
+
+func TestApply_MDMOnly_OverridesDefaults(t *testing.T) {
+	const mdmURL = "https://corp.mdm.example.com:443"
+	withMDMPolicy(t, mdm.NewPolicy(map[string]any{
+		mdm.KeyManagementURL:       mdmURL,
+		mdm.KeyDisableClientRoutes: true,
+		mdm.KeyBlockInbound:        true,
+	}))
+
+	cfg, err := UpdateOrCreateConfig(ConfigInput{
+		ConfigPath: filepath.Join(t.TempDir(), "config.json"),
+	})
+	require.NoError(t, err)
+	require.NotNil(t, cfg)
+
+	assert.Equal(t, mdmURL, cfg.ManagementURL.String())
+	assert.True(t, cfg.DisableClientRoutes)
+	assert.True(t, cfg.BlockInbound)
+
+	assert.True(t, cfg.Policy().HasKey(mdm.KeyManagementURL))
+	assert.True(t, cfg.Policy().HasKey(mdm.KeyDisableClientRoutes))
+	assert.True(t, cfg.Policy().HasKey(mdm.KeyBlockInbound))
+	assert.False(t, cfg.Policy().HasKey(mdm.KeyAllowServerSSH))
+}
+
+func TestApply_MDMBeatsCLIInput(t *testing.T) {
+	const mdmURL = "https://mdm.example.com:443"
+	const cliURL = "https://cli.example.com:443"
+
+	withMDMPolicy(t, mdm.NewPolicy(map[string]any{
+		mdm.KeyManagementURL: mdmURL,
+	}))
+
+	cfg, err := UpdateOrCreateConfig(ConfigInput{
+		ConfigPath:    filepath.Join(t.TempDir(), "config.json"),
+		ManagementURL: cliURL,
+	})
+	require.NoError(t, err)
+	require.NotNil(t, cfg)
+
+	// MDM wins over CLI-supplied management URL.
+	assert.Equal(t, mdmURL, cfg.ManagementURL.String())
+	assert.True(t, cfg.Policy().HasKey(mdm.KeyManagementURL))
+}
+
+func TestApply_MDMInvalidURL_KeepsPreviousValue(t *testing.T) {
+	withMDMPolicy(t, mdm.NewPolicy(map[string]any{
+		mdm.KeyManagementURL: "not-a-url",
+	}))
+
+	cfg, err := UpdateOrCreateConfig(ConfigInput{
+		ConfigPath: filepath.Join(t.TempDir(), "config.json"),
+	})
+	require.NoError(t, err)
+	require.NotNil(t, cfg)
+
+	// Invalid MDM URL is logged and skipped: default URL stays in place
+	// to keep the client functional.
+	assert.Equal(t, DefaultManagementURL, cfg.ManagementURL.String())
+
+	// But the key is still considered MDM-managed (admin intent is to
+	// enforce, daemon rejects user writes to this field — phase-1 scaffolding
+	// reflects this by keeping Policy.HasKey true even on parse failure).
+	assert.True(t, cfg.Policy().HasKey(mdm.KeyManagementURL))
+}
+
+func TestApply_MDMBoolKeysOverrideOnDiskValue(t *testing.T) {
+	tmp := filepath.Join(t.TempDir(), "config.json")
+
+	// Seed without MDM.
+	withMDMPolicy(t, mdm.NewPolicy(nil))
+	_, err := UpdateOrCreateConfig(ConfigInput{
+		ConfigPath:          tmp,
+		DisableClientRoutes: boolPtr(false),
+		RosenpassEnabled:    boolPtr(false),
+	})
+	require.NoError(t, err)
+
+	// Now enable MDM enforcement for these keys.
+	withMDMPolicy(t, mdm.NewPolicy(map[string]any{
+		mdm.KeyDisableClientRoutes: true,
+		mdm.KeyRosenpassEnabled:    true,
+	}))
+
+	cfg, err := UpdateOrCreateConfig(ConfigInput{ConfigPath: tmp})
+	require.NoError(t, err)
+	require.NotNil(t, cfg)
+
+	assert.True(t, cfg.DisableClientRoutes, "MDM override should flip on-disk false to true")
+	assert.True(t, cfg.RosenpassEnabled)
+	assert.True(t, cfg.Policy().HasKey(mdm.KeyDisableClientRoutes))
+	assert.True(t, cfg.Policy().HasKey(mdm.KeyRosenpassEnabled))
+}
+
+func TestApply_MDMPreSharedKeyRedactionSentinelRejected(t *testing.T) {
+	const maskSentinel = "**********"
+
+	withMDMPolicy(t, mdm.NewPolicy(map[string]any{
+		mdm.KeyPreSharedKey: maskSentinel,
+	}))
+
+	cfg, err := UpdateOrCreateConfig(ConfigInput{
+		ConfigPath: filepath.Join(t.TempDir(), "config.json"),
+	})
+	require.NoError(t, err)
+	require.NotNil(t, cfg)
+
+	// Mask sentinel must not be persisted as the actual PSK.
+	assert.NotEqual(t, maskSentinel, cfg.PreSharedKey)
+	// Key still marked managed so user writes are still rejected.
+	assert.True(t, cfg.Policy().HasKey(mdm.KeyPreSharedKey))
+}
+
+func boolPtr(b bool) *bool { return &b }

client/internal/profilemanager/config_test.go

@@ -242,6 +242,35 @@ func TestWireguardPortDefaultVsExplicit(t *testing.T) {
 	}
 }
 
+func TestUpdateConfigServerSSHAllowedNotSet(t *testing.T) {
+	// Configs written before ServerSSHAllowed was introduced lack the field and
+	// unmarshal to nil. Supplying the SSH server flag on top of such a config must
+	// apply the value instead of panicking on a nil pointer dereference.
+	tests := []struct {
+		name  string
+		input *bool
+		want  bool
+	}{
+		{"enable", util.True(), true},
+		{"disable", util.False(), false},
+	}
+
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			configPath := filepath.Join(t.TempDir(), "config.json")
+			require.NoError(t, os.WriteFile(configPath, []byte("{}"), 0600))
+
+			config, err := UpdateConfig(ConfigInput{
+				ConfigPath:       configPath,
+				ServerSSHAllowed: tt.input,
+			})
+			require.NoError(t, err)
+			require.NotNil(t, config.ServerSSHAllowed, "ServerSSHAllowed should be set from input")
+			assert.Equal(t, tt.want, *config.ServerSSHAllowed)
+		})
+	}
+}
+
 func TestUpdateOldManagementURL(t *testing.T) {
 	origProber := newMgmProber
 	newMgmProber = func(_ context.Context, _ string, _ wgtypes.Key, _ bool) (mgmProber, error) {

client/internal/profilemanager/id.go

@@ -0,0 +1,118 @@
+package profilemanager
+
+import (
+	"crypto/rand"
+	"encoding/hex"
+	"fmt"
+	"path/filepath"
+	"strings"
+	"unicode"
+	"unicode/utf8"
+)
+
+const (
+	// profileIDByteLen is the number of random bytes generated for a new
+	// profile ID. The resulting hex string is twice this length.
+	profileIDByteLen = 16
+
+	// shortIDLen is the number of leading characters of an ID we render in
+	// list output. Profiles per device are few, so 8 chars is collision-safe
+	// in practice and easy to type as a prefix.
+	shortIDLen = 8
+
+	// maxProfileNameLen caps the human-readable profile name to keep table
+	// output legible and prevent denial-of-service via huge JSON fields.
+	maxProfileNameLen = 128
+
+	// maxProfileIDLen bounds the on-disk filename we'll accept. New
+	// IDs are 32 hex chars, legacy stems are sanitized profile names. The
+	// cap is generous enough to cover both without permitting absurdly
+	// long filenames.
+	maxProfileIDLen = 64
+)
+
+type ID string
+
+// generateProfileID returns a new random hex ID for a profile file.
+func generateProfileID() (ID, error) {
+	buf := make([]byte, profileIDByteLen)
+	if _, err := rand.Read(buf); err != nil {
+		return "", fmt.Errorf("read random bytes: %w", err)
+	}
+	return ID(hex.EncodeToString(buf)), nil
+}
+
+// IsValidProfileFilenameStem reports whether id is safe to use as the stem
+// of a profile JSON filename.
+func IsValidProfileFilenameStem(id ID) bool {
+	s := id.String()
+	if s == "" || len(s) > maxProfileIDLen {
+		return false
+	}
+	if s == defaultProfileName {
+		return true
+	}
+	if strings.ContainsAny(s, `/\`) || strings.Contains(s, "..") {
+		return false
+	}
+	// filepath.Base catches any leftover separators on platforms with
+	// exotic path conventions.
+	if filepath.Base(s) != s {
+		return false
+	}
+	for _, r := range s {
+		if !(unicode.IsLetter(r) || unicode.IsDigit(r) || r == '_' || r == '-') {
+			return false
+		}
+	}
+	return true
+}
+
+// sanitizeDisplayName normalizes a user-supplied profile display name for
+// storage. It strips ASCII control characters, rejects invalid UTF-8, and
+// caps the length. Emojis, spaces, punctuation, and non-ASCII letters are
+// preserved. Returns an error if nothing usable remains.
+func sanitizeDisplayName(name string) (string, error) {
+	if !utf8.ValidString(name) {
+		return "", fmt.Errorf("name is not valid UTF-8")
+	}
+	name = StripCtrlChars(name)
+	name = strings.TrimSpace(name)
+	if name == "" {
+		return "", fmt.Errorf("name is empty after sanitization")
+	}
+	if utf8.RuneCountInString(name) > maxProfileNameLen {
+		return "", fmt.Errorf("name exceeds %d characters", maxProfileNameLen)
+	}
+	return name, nil
+}
+
+// StripCtrlChars control characters from a name before printing it.
+func StripCtrlChars(name string) string {
+	var b strings.Builder
+	b.Grow(len(name))
+	for _, r := range name {
+		// Skip C0 controls and DEL, plus C1 controls (0x80–0x9F).
+		if r < 0x20 || r == 0x7F || (r >= 0x80 && r <= 0x9F) {
+			continue
+		}
+		b.WriteRune(r)
+	}
+	return b.String()
+}
+
+// ShortID truncates an ID for display.
+func (id ID) ShortID() string {
+	if id == DefaultProfileName {
+		return DefaultProfileName
+	}
+	runes := []rune(id)
+	if len(runes) <= shortIDLen {
+		return id.String()
+	}
+	return string(runes[:shortIDLen])
+}
+
+func (id ID) String() string {
+	return string(id)
+}

client/internal/profilemanager/profilemanager.go

@@ -19,19 +19,41 @@ const (
 )
 
 type Profile struct {
-	Name     string
+	// ID is the on-disk filename stem (without .json). For new profiles
+	// it is a 32-char hex string; legacy profiles created before the
+	// ID-keyed layout keep their original name as their ID. The reserved
+	// value "default" identifies the special default profile.
+	ID ID
+	// Name is the human-readable display name. Falls back to ID when the
+	// underlying JSON has no "name" field set.
+	Name string
+	// Path is the absolute path to the profile JSON. Populated by the
+	// loader so callers do not have to reconstruct it from ID + dir.
+	Path     string
 	IsActive bool
 }
 
 func (p *Profile) FilePath() (string, error) {
-	if p.Name == "" {
+	if p.Path != "" {
-		return "", fmt.Errorf("active profile name is empty")
+		return p.Path, nil
 	}
 
-	if p.Name == defaultProfileName {
+	id := p.ID
+	if id == "" {
+		id = ID(p.Name)
+	}
+	if id == "" {
+		return "", fmt.Errorf("profile ID is empty")
+	}
+
+	if id == defaultProfileName {
 		return DefaultConfigPath, nil
 	}
 
+	if !IsValidProfileFilenameStem(id) {
+		return "", fmt.Errorf("invalid profile ID: %q", id)
+	}
+
 	username, err := user.Current()
 	if err != nil {
 		return "", fmt.Errorf("failed to get current user: %w", err)
@@ -42,10 +64,13 @@ func (p *Profile) FilePath() (string, error) {
 		return "", fmt.Errorf("failed to get config directory for user %s: %w", username.Username, err)
 	}
 
-	return filepath.Join(configDir, p.Name+".json"), nil
+	return filepath.Join(configDir, id.String()+".json"), nil
 }
 
 func (p *Profile) IsDefault() bool {
+	if p.ID != "" {
+		return p.ID == defaultProfileName
+	}
 	return p.Name == defaultProfileName
 }
 
@@ -57,18 +82,24 @@ func NewProfileManager() *ProfileManager {
 	return &ProfileManager{}
 }
 
+// GetActiveProfile returns the active profile as recorded in the local
+// user state file. Only ID is populated.
 func (pm *ProfileManager) GetActiveProfile() (*Profile, error) {
 	pm.mu.Lock()
 	defer pm.mu.Unlock()
 
-	prof := pm.getActiveProfileState()
+	id := pm.getActiveProfileState()
-	return &Profile{Name: prof}, nil
+	return &Profile{ID: id}, nil
 }
 
-func (pm *ProfileManager) SwitchProfile(profileName string) error {
+// SwitchProfile records the given profile ID as active in the local user
-	profileName = sanitizeProfileName(profileName)
+// state file.
+func (pm *ProfileManager) SwitchProfile(id ID) error {
+	if id != defaultProfileName && !IsValidProfileFilenameStem(id) {
+		return fmt.Errorf("invalid profile ID: %q", id)
+	}
 
-	if err := pm.setActiveProfileState(profileName); err != nil {
+	if err := pm.setActiveProfileState(id); err != nil {
 		return fmt.Errorf("failed to switch profile: %w", err)
 	}
 	return nil
@@ -85,7 +116,7 @@ func sanitizeProfileName(name string) string {
 	}, name)
 }
 
-func (pm *ProfileManager) getActiveProfileState() string {
+func (pm *ProfileManager) getActiveProfileState() ID {
 
 	configDir, err := getConfigDir()
 	if err != nil {
@@ -113,10 +144,10 @@ func (pm *ProfileManager) getActiveProfileState() string {
 		return defaultProfileName
 	}
 
-	return profileName
+	return ID(profileName)
 }
 
-func (pm *ProfileManager) setActiveProfileState(profileName string) error {
+func (pm *ProfileManager) setActiveProfileState(id ID) error {
 
 	configDir, err := getConfigDir()
 	if err != nil {
@@ -125,7 +156,7 @@ func (pm *ProfileManager) setActiveProfileState(profileName string) error {
 
 	statePath := filepath.Join(configDir, activeProfileStateFilename)
 
-	err = os.WriteFile(statePath, []byte(profileName), 0600)
+	err = os.WriteFile(statePath, []byte(id), 0600)
 	if err != nil {
 		return fmt.Errorf("failed to write active profile state: %w", err)
 	}
@@ -142,7 +173,7 @@ func GetLoginHint() string {
 		return ""
 	}
 
-	profileState, err := pm.GetProfileState(activeProf.Name)
+	profileState, err := pm.GetProfileState(activeProf.ID)
 	if err != nil {
 		log.Debugf("failed to get profile state for login hint: %v", err)
 		return ""

client/internal/profilemanager/profilemanager_test.go

@@ -50,14 +50,14 @@ func TestServiceManager_CreateAndGetDefaultProfile(t *testing.T) {
 
 			state, err := sm.GetActiveProfileState()
 			assert.NoError(t, err)
-			assert.Equal(t, state.Name, defaultProfileName) // No active profile state yet
+			assert.Equal(t, defaultProfileName, state.ID.String()) // No active profile state yet
 
 			err = sm.SetActiveProfileStateToDefault()
 			assert.NoError(t, err)
 
 			active, err := sm.GetActiveProfileState()
 			assert.NoError(t, err)
-			assert.Equal(t, "default", active.Name)
+			assert.Equal(t, "default", active.ID.String())
 		})
 	})
 }
@@ -92,14 +92,14 @@ func TestServiceManager_SetActiveProfileState(t *testing.T) {
 			currUser, err := user.Current()
 			assert.NoError(t, err)
 			sm := &ServiceManager{}
-			state := &ActiveProfileState{Name: "foo", Username: currUser.Username}
+			state := &ActiveProfileState{ID: "foo", Username: currUser.Username}
 			err = sm.SetActiveProfileState(state)
 			assert.NoError(t, err)
 
 			// Should error on nil or incomplete state
 			err = sm.SetActiveProfileState(nil)
 			assert.Error(t, err)
-			err = sm.SetActiveProfileState(&ActiveProfileState{Name: "", Username: ""})
+			err = sm.SetActiveProfileState(&ActiveProfileState{ID: "", Username: ""})
 			assert.Error(t, err)
 		})
 	})

client/internal/profilemanager/service.go

@@ -2,6 +2,7 @@ package profilemanager
 
 import (
 	"context"
+	"encoding/json"
 	"errors"
 	"fmt"
 	"io"
@@ -23,12 +24,43 @@ var (
 	DefaultConfigPathDir   = ""
 	DefaultConfigPath      = ""
 	ActiveProfileStatePath = ""
-)
 
-var (
 	ErrorOldDefaultConfigNotFound = errors.New("old default config not found")
 )
 
+// ErrAmbiguousHandle is returned when a profile handle (ID prefix or name)
+// matches more than one profile. Callers can render Candidates to help the
+// user disambiguate.
+type ErrAmbiguousHandle struct {
+	Handle     string
+	Candidates []Profile
+	Kind       AmbiguityKind
+}
+
+// AmbiguityKind describes which matcher produced the ambiguity, so callers
+// can tailor the error message.
+type AmbiguityKind int
+
+const (
+	AmbiguityKindIDPrefix AmbiguityKind = iota
+	AmbiguityKindName
+)
+
+// profileMeta is the minimal slice of a profile JSON we need, so we avoid
+// reading all fields
+type profileMeta struct {
+	Name string
+}
+
+func (e *ErrAmbiguousHandle) Error() string {
+	switch e.Kind {
+	case AmbiguityKindIDPrefix:
+		return fmt.Sprintf("ID prefix %q is ambiguous (matches %d profiles)", e.Handle, len(e.Candidates))
+	default:
+		return fmt.Sprintf("name %q is ambiguous (%d profiles share this name)", e.Handle, len(e.Candidates))
+	}
+}
+
 func init() {
 
 	DefaultConfigPathDir = "/var/lib/netbird/"
@@ -54,25 +86,34 @@ func init() {
 }
 
 type ActiveProfileState struct {
-	Name     string `json:"name"`
+	// ID is the on-disk filename stem of the active profile. The JSON tag stays
+	// as "name" for backwards compatibility with active state files written
+	// before the ID-based config files. Legacy values were profile names, which
+	// were also the legacy filename stems, so they still resolve to the correct
+	// file on disk.
+	ID       ID     `json:"name"`
 	Username string `json:"username"`
 }
 
 func (a *ActiveProfileState) FilePath() (string, error) {
-	if a.Name == "" {
+	if a.ID == "" {
-		return "", fmt.Errorf("active profile name is empty")
+		return "", fmt.Errorf("active profile ID is empty")
 	}
 
-	if a.Name == defaultProfileName {
+	if a.ID == defaultProfileName {
 		return DefaultConfigPath, nil
 	}
 
+	if !IsValidProfileFilenameStem(a.ID) {
+		return "", fmt.Errorf("invalid profile ID: %q", a.ID)
+	}
+
 	configDir, err := getConfigDirForUser(a.Username)
 	if err != nil {
 		return "", fmt.Errorf("failed to get config directory for user %s: %w", a.Username, err)
 	}
 
-	return filepath.Join(configDir, a.Name+".json"), nil
+	return filepath.Join(configDir, a.ID.String()+".json"), nil
 }
 
 type ServiceManager struct {
@@ -178,7 +219,7 @@ func (s *ServiceManager) GetActiveProfileState() (*ActiveProfileState, error) {
 				return nil, fmt.Errorf("failed to set active profile to default: %w", err)
 			}
 			return &ActiveProfileState{
-				Name:     "default",
+				ID:       defaultProfileName,
 				Username: "",
 			}, nil
 		} else {
@@ -186,12 +227,12 @@ func (s *ServiceManager) GetActiveProfileState() (*ActiveProfileState, error) {
 		}
 	}
 
-	if activeProfile.Name == "" {
+	if activeProfile.ID == "" {
 		if err := s.SetActiveProfileStateToDefault(); err != nil {
 			return nil, fmt.Errorf("failed to set active profile to default: %w", err)
 		}
 		return &ActiveProfileState{
-			Name:     "default",
+			ID:       defaultProfileName,
 			Username: "",
 		}, nil
 	}
@@ -216,25 +257,29 @@ func (s *ServiceManager) setDefaultActiveState() error {
 }
 
 func (s *ServiceManager) SetActiveProfileState(a *ActiveProfileState) error {
-	if a == nil || a.Name == "" {
+	if a == nil || a.ID == "" {
 		return errors.New("invalid active profile state")
 	}
 
-	if a.Name != defaultProfileName && a.Username == "" {
+	if a.ID != defaultProfileName && a.Username == "" {
-		return fmt.Errorf("username must be set for non-default profiles, got: %s", a.Name)
+		return fmt.Errorf("username must be set for non-default profiles, got: %s", a.ID)
+	}
+
+	if a.ID != defaultProfileName && !IsValidProfileFilenameStem(a.ID) {
+		return fmt.Errorf("invalid profile ID: %q", a.ID)
 	}
 
 	if err := util.WriteJsonWithRestrictedPermission(context.Background(), ActiveProfileStatePath, a); err != nil {
 		return fmt.Errorf("failed to write active profile state: %w", err)
 	}
 
-	log.Infof("active profile set to %s for %s", a.Name, a.Username)
+	log.Infof("active profile set to %s for %s", a.ID, a.Username)
 	return nil
 }
 
 func (s *ServiceManager) SetActiveProfileStateToDefault() error {
 	return s.SetActiveProfileState(&ActiveProfileState{
-		Name:     "default",
+		ID:       defaultProfileName,
 		Username: "",
 	})
 }
@@ -243,57 +288,117 @@ func (s *ServiceManager) DefaultProfilePath() string {
 	return DefaultConfigPath
 }
 
-func (s *ServiceManager) AddProfile(profileName, username string) error {
+// AddProfile creates a new profile with a generated ID. The user-supplied
+// displayName is stored inside the JSON's name field, the on-disk filename
+// uses the generated ID.
+//
+// The returned Profile carries the freshly-generated ID so callers can
+// show it to the user (and so the gRPC AddProfileResponse can include
+// it).
+func (s *ServiceManager) AddProfile(displayName, username string) (*Profile, error) {
 	configDir, err := s.getConfigDir(username)
 	if err != nil {
-		return fmt.Errorf("failed to get config directory: %w", err)
+		return nil, fmt.Errorf("failed to get config directory: %w", err)
 	}
 
-	profileName = sanitizeProfileName(profileName)
+	displayName, err = sanitizeDisplayName(displayName)
-
-	if profileName == defaultProfileName {
-		return fmt.Errorf("cannot create profile with reserved name: %s", defaultProfileName)
-	}
-
-	profPath := filepath.Join(configDir, profileName+".json")
-	profileExists, err := fileExists(profPath)
 	if err != nil {
-		return fmt.Errorf("failed to check if profile exists: %w", err)
+		return nil, fmt.Errorf("invalid profile name: %w", err)
-	}
-	if profileExists {
-		return ErrProfileAlreadyExists
 	}
 
+	id, err := generateProfileID()
+	if err != nil {
+		return nil, fmt.Errorf("generate profile id: %w", err)
+	}
+
+	profPath := filepath.Join(configDir, id.String()+".json")
 	cfg, err := createNewConfig(ConfigInput{ConfigPath: profPath})
 	if err != nil {
-		return fmt.Errorf("failed to create new config: %w", err)
+		return nil, fmt.Errorf("failed to create new config: %w", err)
+	}
+	cfg.Name = displayName
+
+	if err := util.WriteJson(context.Background(), profPath, cfg); err != nil {
+		return nil, fmt.Errorf("failed to write profile config: %w", err)
 	}
 
-	err = util.WriteJson(context.Background(), profPath, cfg)
+	return &Profile{
+		ID:   id,
+		Name: displayName,
+		Path: profPath,
+	}, nil
+}
+
+func (s *ServiceManager) RenameProfile(id ID, username string, newName string) error {
+	displayName, err := sanitizeDisplayName(newName)
 	if err != nil {
-		return fmt.Errorf("failed to write profile config: %w", err)
+		return fmt.Errorf("invalid profile name: %w", err)
 	}
 
+	if !IsValidProfileFilenameStem(id) {
+		return fmt.Errorf("invalid profile ID: %q", id)
+	}
+
+	profiles, err := s.loadAllProfiles(username)
+	if err != nil {
+		return fmt.Errorf("load profiles: %w", err)
+	}
+
+	var target *Profile
+	for i := range profiles {
+		if profiles[i].ID == id {
+			target = &profiles[i]
+			break
+		}
+	}
+	if target == nil {
+		return ErrProfileNotFound
+	}
+
+	data, err := os.ReadFile(target.Path)
+	if err != nil {
+		return err
+	}
+	var cfg Config
+	if err := json.Unmarshal(data, &cfg); err != nil {
+		return err
+	}
+	cfg.Name = displayName
+
+	if err := util.WriteJson(context.Background(), target.Path, cfg); err != nil {
+		return fmt.Errorf("failed to write profile name: %w", err)
+	}
 	return nil
 }
 
-func (s *ServiceManager) RemoveProfile(profileName, username string) error {
+// RemoveProfile deletes the profile identified by id. Callers must have
-	configDir, err := s.getConfigDir(username)
+// already resolved any user-supplied handle to a concrete ID via
-	if err != nil {
+// ResolveProfile.
-		return fmt.Errorf("failed to get config directory: %w", err)
+func (s *ServiceManager) RemoveProfile(id ID, username string) error {
+	if id == defaultProfileName {
+		defaultName := readProfileName(DefaultConfigPath)
+		if defaultName == "" {
+			defaultName = defaultProfileName
+		}
+		return fmt.Errorf("cannot remove default profile with name: %s", defaultName)
+	}
+	if !IsValidProfileFilenameStem(id) {
+		return fmt.Errorf("invalid profile ID: %q", id)
 	}
 
-	profileName = sanitizeProfileName(profileName)
+	profiles, err := s.loadAllProfiles(username)
-
-	if profileName == defaultProfileName {
-		return fmt.Errorf("cannot remove profile with reserved name: %s", defaultProfileName)
-	}
-	profPath := filepath.Join(configDir, profileName+".json")
-	profileExists, err := fileExists(profPath)
 	if err != nil {
-		return fmt.Errorf("failed to check if profile exists: %w", err)
+		return fmt.Errorf("load profiles: %w", err)
 	}
-	if !profileExists {
+
+	var target *Profile
+	for i := range profiles {
+		if profiles[i].ID == id {
+			target = &profiles[i]
+			break
+		}
+	}
+	if target == nil {
 		return ErrProfileNotFound
 	}
 
@@ -301,57 +406,26 @@ func (s *ServiceManager) RemoveProfile(profileName, username string) error {
 	if err != nil && !errors.Is(err, ErrNoActiveProfile) {
 		return fmt.Errorf("failed to get active profile: %w", err)
 	}
-
+	if activeProf != nil && activeProf.ID == id {
-	if activeProf != nil && activeProf.Name == profileName {
+		return fmt.Errorf("cannot remove active profile: %s", id)
-		return fmt.Errorf("cannot remove active profile: %s", profileName)
 	}
 
-	err = util.RemoveJson(profPath)
+	if err := util.RemoveJson(target.Path); err != nil {
-	if err != nil {
 		return fmt.Errorf("failed to remove profile config: %w", err)
 	}
+
+	stateFile := filepath.Join(filepath.Dir(target.Path), id.String()+".state.json")
+	if err := os.Remove(stateFile); err != nil && !os.IsNotExist(err) {
+		log.Warnf("failed to remove profile state file %s: %v", stateFile, err)
+	}
+
 	return nil
 }
 
+// ListProfiles returns every profile for the given user, including the
+// default profile, with IsActive flags set.
 func (s *ServiceManager) ListProfiles(username string) ([]Profile, error) {
-	configDir, err := s.getConfigDir(username)
+	return s.loadAllProfiles(username)
-	if err != nil {
-		return nil, fmt.Errorf("failed to get config directory: %w", err)
-	}
-
-	files, err := util.ListFiles(configDir, "*.json")
-	if err != nil {
-		return nil, fmt.Errorf("failed to list profile files: %w", err)
-	}
-
-	var filtered []string
-	for _, file := range files {
-		if strings.HasSuffix(file, "state.json") {
-			continue // skip state files
-		}
-		filtered = append(filtered, file)
-	}
-	sort.Strings(filtered)
-
-	var activeProfName string
-	activeProf, err := s.GetActiveProfileState()
-	if err == nil {
-		activeProfName = activeProf.Name
-	}
-
-	var profiles []Profile
-	// add default profile always
-	profiles = append(profiles, Profile{Name: defaultProfileName, IsActive: activeProfName == "" || activeProfName == defaultProfileName})
-	for _, file := range filtered {
-		profileName := strings.TrimSuffix(filepath.Base(file), ".json")
-		var isActive bool
-		if activeProfName != "" && activeProfName == profileName {
-			isActive = true
-		}
-		profiles = append(profiles, Profile{Name: profileName, IsActive: isActive})
-	}
-
-	return profiles, nil
 }
 
 // GetStatePath returns the path to the state file based on the operating system
@@ -369,7 +443,12 @@ func (s *ServiceManager) GetStatePath() string {
 		return defaultStatePath
 	}
 
-	if activeProf.Name == defaultProfileName {
+	if activeProf.ID == defaultProfileName {
+		return defaultStatePath
+	}
+
+	if !IsValidProfileFilenameStem(activeProf.ID) {
+		log.Warnf("invalid active profile ID %q, using default state path", activeProf.ID)
 		return defaultStatePath
 	}
 
@@ -379,7 +458,7 @@ func (s *ServiceManager) GetStatePath() string {
 		return defaultStatePath
 	}
 
-	return filepath.Join(configDir, activeProf.Name+".state.json")
+	return filepath.Join(configDir, activeProf.ID.String()+".state.json")
 }
 
 // getConfigDir returns the profiles directory, using profilesDir if set, otherwise getConfigDirForUser
@@ -390,3 +469,169 @@ func (s *ServiceManager) getConfigDir(username string) (string, error) {
 
 	return getConfigDirForUser(username)
 }
+
+// loadAllProfiles returns every profile visible to the daemon for the
+// given user, including the default profile. The returned slice is sorted
+// by ID for a stable display order.
+//
+// Each Profile is fully populated: ID is the filename stem, Name comes
+// from the JSON's "name" field (falling back to the filename stem when absent)
+// and Path is built from a basename read off disk.
+func (s *ServiceManager) loadAllProfiles(username string) ([]Profile, error) {
+	activeID, activeIsDefault := s.activeProfileID()
+	defaultName := readProfileName(DefaultConfigPath)
+	if defaultName == "" {
+		defaultName = defaultProfileName
+	}
+
+	profiles := []Profile{{
+		ID:       defaultProfileName,
+		Name:     defaultName,
+		Path:     DefaultConfigPath,
+		IsActive: activeIsDefault,
+	}}
+
+	configDir, err := s.getConfigDir(username)
+	if err != nil {
+		return nil, fmt.Errorf("get config directory: %w", err)
+	}
+
+	entries, err := os.ReadDir(configDir)
+	if err != nil {
+		if errors.Is(err, os.ErrNotExist) {
+			return profiles, nil
+		}
+		return nil, fmt.Errorf("read profile directory: %w", err)
+	}
+
+	var fileProfiles []Profile
+	for _, entry := range entries {
+		if entry.IsDir() {
+			continue
+		}
+		base := entry.Name()
+		if !strings.HasSuffix(base, ".json") {
+			continue
+		}
+		if strings.HasSuffix(base, ".state.json") {
+			continue
+		}
+		stem := ID(strings.TrimSuffix(base, ".json"))
+		if stem == defaultProfileName {
+			// default lives at the top-level config dir, not under /<user>
+			continue
+		}
+		if !IsValidProfileFilenameStem(ID(stem)) {
+			continue
+		}
+		path := filepath.Join(configDir, base)
+		name := readProfileName(path)
+		if name == "" {
+			name = stem.String()
+		}
+		fileProfiles = append(fileProfiles, Profile{
+			ID:       stem,
+			Name:     name,
+			Path:     path,
+			IsActive: stem == ID(activeID),
+		})
+	}
+
+	sort.Slice(fileProfiles, func(i, j int) bool {
+		if fileProfiles[i].Name != fileProfiles[j].Name {
+			return fileProfiles[i].Name < fileProfiles[j].Name
+		}
+		// Sort tie-break on ID so duplicate names always render in the same order.
+		return fileProfiles[i].ID < fileProfiles[j].ID
+	})
+	profiles = append(profiles, fileProfiles...)
+	return profiles, nil
+}
+
+// readProfileName parses just the "name" field from the profile Json.
+func readProfileName(path string) string {
+	data, err := os.ReadFile(path)
+	if err != nil {
+		return ""
+	}
+	var meta profileMeta
+	if err := json.Unmarshal(data, &meta); err != nil {
+		return ""
+	}
+	return meta.Name
+}
+
+// activeProfileID returns the currently-active profile's ID. The second
+// return value is true when the active profile is the default one.
+func (s *ServiceManager) activeProfileID() (ID, bool) {
+	state, err := s.GetActiveProfileState()
+	if err != nil || state == nil {
+		return defaultProfileName, true
+	}
+	if state.ID == "" || state.ID == defaultProfileName {
+		return defaultProfileName, true
+	}
+	return state.ID, false
+}
+
+// ResolveProfile turns a user-supplied handle into a Profile. Resolution
+// precedence is: exact ID match, then unique exact name, then unique ID
+// prefix. Ambiguous matches return *ErrAmbiguousHandle so callers can
+// surface the candidates.
+func (s *ServiceManager) ResolveProfile(handle, username string) (*Profile, error) {
+	if handle == "" {
+		return nil, fmt.Errorf("profile handle is empty")
+	}
+
+	profiles, err := s.loadAllProfiles(username)
+	if err != nil {
+		return nil, err
+	}
+
+	for i := range profiles {
+		if profiles[i].ID == ID(handle) {
+			return &profiles[i], nil
+		}
+	}
+
+	var nameMatches []Profile
+	for i := range profiles {
+		if profiles[i].Name == handle {
+			nameMatches = append(nameMatches, profiles[i])
+		}
+	}
+	if len(nameMatches) == 1 {
+		return &nameMatches[0], nil
+	}
+	if len(nameMatches) > 1 {
+		return nil, &ErrAmbiguousHandle{
+			Handle:     handle,
+			Candidates: nameMatches,
+			Kind:       AmbiguityKindName,
+		}
+	}
+
+	// ID prefix match. Skip the default profile so `select d` does not
+	// accidentally pick it via prefix.
+	var prefixMatches []Profile
+	for i := range profiles {
+		if profiles[i].ID == defaultProfileName {
+			continue
+		}
+		if strings.HasPrefix(profiles[i].ID.String(), handle) {
+			prefixMatches = append(prefixMatches, profiles[i])
+		}
+	}
+	if len(prefixMatches) == 1 {
+		return &prefixMatches[0], nil
+	}
+	if len(prefixMatches) > 1 {
+		return nil, &ErrAmbiguousHandle{
+			Handle:     handle,
+			Candidates: prefixMatches,
+			Kind:       AmbiguityKindIDPrefix,
+		}
+	}
+
+	return nil, ErrProfileNotFound
+}

client/internal/profilemanager/service_test.go

@@ -0,0 +1,230 @@
+package profilemanager
+
+import (
+	"context"
+	"errors"
+	"os"
+	"os/user"
+	"path/filepath"
+	"strings"
+	"testing"
+
+	"github.com/stretchr/testify/assert"
+	"github.com/stretchr/testify/require"
+
+	"github.com/netbirdio/netbird/util"
+)
+
+// withTestSM wires up patched globals + a clean config dir and returns a
+// fully initialized ServiceManager plus the username we are scoped to.
+func withTestSM(t *testing.T, fn func(sm *ServiceManager, username string)) {
+	t.Helper()
+	withTempConfigDir(t, func(configDir string) {
+		withPatchedGlobals(t, configDir, func() {
+			u, err := user.Current()
+			require.NoError(t, err)
+			sm := &ServiceManager{}
+			require.NoError(t, sm.CreateDefaultProfile())
+			fn(sm, u.Username)
+		})
+	})
+}
+
+func TestServiceProfile_ExactID(t *testing.T) {
+	withTestSM(t, func(sm *ServiceManager, username string) {
+		created, err := sm.AddProfile("work", username)
+		require.NoError(t, err)
+
+		got, err := sm.ResolveProfile(created.ID.String(), username)
+		require.NoError(t, err)
+		assert.Equal(t, created.ID, got.ID)
+		assert.Equal(t, "work", got.Name)
+	})
+}
+
+func TestServiceProfile_IDPrefix(t *testing.T) {
+	withTestSM(t, func(sm *ServiceManager, username string) {
+		created, err := sm.AddProfile("work", username)
+		require.NoError(t, err)
+
+		prefix := created.ID[:4]
+		got, err := sm.ResolveProfile(prefix.String(), username)
+		require.NoError(t, err)
+		assert.Equal(t, created.ID, got.ID)
+	})
+}
+
+func TestServiceProfile_AmbiguousPrefix(t *testing.T) {
+	withTestSM(t, func(sm *ServiceManager, username string) {
+		// Plant two profiles whose IDs share a known prefix by writing
+		// the files directly, since generated IDs are random.
+		configDir, err := sm.getConfigDir(username)
+		require.NoError(t, err)
+		for _, id := range []string{"abcd1111aaaa", "abcd2222bbbb"} {
+			path := filepath.Join(configDir, id+".json")
+			require.NoError(t, util.WriteJson(context.Background(), path, &Config{Name: id}))
+		}
+
+		_, err = sm.ResolveProfile("abcd", username)
+		var amb *ErrAmbiguousHandle
+		require.ErrorAs(t, err, &amb)
+		assert.Equal(t, AmbiguityKindIDPrefix, amb.Kind)
+		assert.Len(t, amb.Candidates, 2)
+	})
+}
+
+func TestServiceProfile_ExactNameUnique(t *testing.T) {
+	withTestSM(t, func(sm *ServiceManager, username string) {
+		_, err := sm.AddProfile("work", username)
+		require.NoError(t, err)
+
+		got, err := sm.ResolveProfile("work", username)
+		require.NoError(t, err)
+		assert.Equal(t, "work", got.Name)
+	})
+}
+
+func TestServiceProfile_AmbiguousName(t *testing.T) {
+	withTestSM(t, func(sm *ServiceManager, username string) {
+		_, err := sm.AddProfile("work", username)
+		require.NoError(t, err)
+		_, err = sm.AddProfile("work", username)
+		require.NoError(t, err)
+
+		_, err = sm.ResolveProfile("work", username)
+		var amb *ErrAmbiguousHandle
+		require.ErrorAs(t, err, &amb)
+		assert.Equal(t, AmbiguityKindName, amb.Kind)
+		assert.Len(t, amb.Candidates, 2)
+	})
+}
+
+func TestServiceProfile_NotFound(t *testing.T) {
+	withTestSM(t, func(sm *ServiceManager, username string) {
+		_, err := sm.ResolveProfile("nope", username)
+		assert.ErrorIs(t, err, ErrProfileNotFound)
+	})
+}
+
+func TestServiceProfile_DefaultByExactID(t *testing.T) {
+	withTestSM(t, func(sm *ServiceManager, username string) {
+		got, err := sm.ResolveProfile(defaultProfileName, username)
+		require.NoError(t, err)
+		assert.Equal(t, defaultProfileName, got.ID.String())
+	})
+}
+
+func TestServiceProfile_LegacyFilenameCoexists(t *testing.T) {
+	// Legacy profiles stored as <name>.json with no "name" JSON field
+	// should still be discoverable by name and removable by name.
+	withTestSM(t, func(sm *ServiceManager, username string) {
+		configDir, err := sm.getConfigDir(username)
+		require.NoError(t, err)
+		path := filepath.Join(configDir, "legacy.json")
+		require.NoError(t, util.WriteJson(context.Background(), path, &Config{}))
+
+		got, err := sm.ResolveProfile("legacy", username)
+		require.NoError(t, err)
+		assert.Equal(t, "legacy", got.ID.String())
+		// Name falls back to the filename stem when JSON omits it.
+		assert.Equal(t, "legacy", got.Name)
+	})
+}
+
+func TestAddProfile_AllowsDuplicateWithFlag(t *testing.T) {
+	withTestSM(t, func(sm *ServiceManager, username string) {
+		first, err := sm.AddProfile("work", username)
+		require.NoError(t, err)
+
+		second, err := sm.AddProfile("work", username)
+		require.NoError(t, err)
+		assert.NotEqual(t, first.ID, second.ID)
+		assert.Equal(t, "work", second.Name)
+	})
+}
+
+func TestAddProfile_RejectsInvalidNames(t *testing.T) {
+	withTestSM(t, func(sm *ServiceManager, username string) {
+		cases := []string{
+			"",                                       // empty
+			"\x00\x01",                               // only control chars (becomes empty)
+			strings.Repeat("a", maxProfileNameLen+1), // too long
+		}
+		for _, name := range cases {
+			_, err := sm.AddProfile(name, username)
+			assert.Error(t, err, "expected error for %q", name)
+		}
+	})
+}
+
+func TestRemoveProfile_RejectsInvalidID(t *testing.T) {
+	withTestSM(t, func(sm *ServiceManager, username string) {
+		err := sm.RemoveProfile("../escape", username)
+		assert.Error(t, err)
+	})
+}
+
+func TestSanitizeDisplayName(t *testing.T) {
+	cases := []struct {
+		in      string
+		want    string
+		wantErr bool
+	}{
+		{"work", "work", false},
+		{"My Work Account", "My Work Account", false},
+		{"emoji 🚀 ok", "emoji 🚀 ok", false},
+		{"漢字テスト", "漢字テスト", false},
+		{"with\x00null", "withnull", false},
+		{"\x01\x02\x03", "", true},
+		{"", "", true},
+	}
+	for _, tc := range cases {
+		got, err := sanitizeDisplayName(tc.in)
+		if tc.wantErr {
+			assert.Error(t, err, "case %q", tc.in)
+			continue
+		}
+		assert.NoError(t, err, "case %q", tc.in)
+		assert.Equal(t, tc.want, got, "case %q", tc.in)
+	}
+}
+
+func TestIsValidProfileFilenameStem(t *testing.T) {
+	cases := []struct {
+		in   string
+		want bool
+	}{
+		{"default", true},
+		{"abc123def456", true},
+		{"legacy-name", true},
+		{"legacy_name", true},
+		{"", false},
+		{"..", false},
+		{"../etc", false},
+		{"foo/bar", false},
+		{`foo\bar`, false},
+		{"with space", false},
+		{"with.dot", false},
+		{strings.Repeat("a", maxProfileIDLen+1), false},
+	}
+	for _, tc := range cases {
+		got := IsValidProfileFilenameStem(ID(tc.in))
+		assert.Equal(t, tc.want, got, "case %q", tc.in)
+	}
+}
+
+func TestRemoveProfile_DeletesStateFile(t *testing.T) {
+	withTestSM(t, func(sm *ServiceManager, username string) {
+		created, err := sm.AddProfile("work", username)
+		require.NoError(t, err)
+
+		configDir, err := sm.getConfigDir(username)
+		require.NoError(t, err)
+		statePath := filepath.Join(configDir, created.ID.String()+".state.json")
+		require.NoError(t, os.WriteFile(statePath, []byte(`{"email":"a@b"}`), 0600))
+
+		require.NoError(t, sm.RemoveProfile(created.ID, username))
+		_, err = os.Stat(statePath)
+		assert.True(t, errors.Is(err, os.ErrNotExist), "state file should be removed")
+	})
+}

client/internal/profilemanager/state.go

@@ -13,13 +13,20 @@ type ProfileState struct {
 	Email string `json:"email"`
 }
 
-func (pm *ProfileManager) GetProfileState(profileName string) (*ProfileState, error) {
+// GetProfileState reads the per-profile state file keyed by profile ID.
+// The state file lives in the user's config directory. Legacy state files
+// keyed by the old profile name remain readable.
+func (pm *ProfileManager) GetProfileState(id ID) (*ProfileState, error) {
 	configDir, err := getConfigDir()
 	if err != nil {
 		return nil, fmt.Errorf("get config directory: %w", err)
 	}
 
-	stateFile := filepath.Join(configDir, profileName+".state.json")
+	if id != defaultProfileName && !IsValidProfileFilenameStem(id) {
+		return nil, fmt.Errorf("invalid profile ID: %q", id)
+	}
+
+	stateFile := filepath.Join(configDir, id.String()+".state.json")
 	stateFileExists, err := fileExists(stateFile)
 	if err != nil {
 		return nil, fmt.Errorf("failed to check if profile state file exists: %w", err)
@@ -51,7 +58,12 @@ func (pm *ProfileManager) SetActiveProfileState(state *ProfileState) error {
 		return fmt.Errorf("get active profile: %w", err)
 	}
 
-	stateFile := filepath.Join(configDir, activeProf.Name+".state.json")
+	id := activeProf.ID
+	if id != defaultProfileName && !IsValidProfileFilenameStem(id) {
+		return fmt.Errorf("invalid active profile ID: %q", id)
+	}
+
+	stateFile := filepath.Join(configDir, id.String()+".state.json")
 	err = util.WriteJsonWithRestrictedPermission(context.Background(), stateFile, state)
 	if err != nil {
 		return fmt.Errorf("write profile state: %w", err)

client/internal/relay/relay.go

@@ -32,6 +32,9 @@ type ProbeResult struct {
 	URI  string
 	Err  error
 	Addr string
+	// Transport is the negotiated relay transport, empty
+	// for stun/turn probes or when not connected.
+	Transport string
 }
 
 type StunTurnProbe struct {

client/internal/rosenpass/manager_test.go

@@ -22,14 +22,14 @@ type removePeerCall struct {
 }
 
 type mockServer struct {
-	mu         sync.Mutex
+	mu        sync.Mutex
-	addCalls   []addPeerCall
+	addCalls  []addPeerCall
-	removed    []removePeerCall
+	removed   []removePeerCall
-	nextID     rp.PeerID
+	nextID    rp.PeerID
-	addErr     error
+	addErr    error
-	removeErr  error
+	removeErr error
-	closed     bool
+	closed    bool
-	ran        bool
+	ran       bool
 }
 
 func (m *mockServer) AddPeer(cfg rp.PeerConfig) (rp.PeerID, error) {
@@ -51,7 +51,7 @@ func (m *mockServer) RemovePeer(id rp.PeerID) error {
 	return m.removeErr
 }
 
-func (m *mockServer) Run() error  { m.ran = true; return nil }
+func (m *mockServer) Run() error   { m.ran = true; return nil }
 func (m *mockServer) Close() error { m.closed = true; return nil }
 
 type setPSKCall struct {

client/internal/rosenpass/seed_test.go

@@ -41,4 +41,3 @@ func TestDeterministicSeedKey_TooShortKey_ReturnsError(t *testing.T) {
 	_, err = DeterministicSeedKey(long, short)
 	require.Error(t, err)
 }
-

client/internal/routemanager/dnsinterceptor/handler.go

@@ -251,6 +251,14 @@ func (d *DnsInterceptor) ServeDNS(w dns.ResponseWriter, r *dns.Msg) {
 		r.MsgHdr.AuthenticatedData = true
 	}
 
+	// Advertise EDNS0 to the forwarder so it may return an Extended DNS Error
+	// describing why a lookup failed. The OPT is stripped from the reply when
+	// the original client did not request EDNS0.
+	hadEdns := r.IsEdns0() != nil
+	if !hadEdns {
+		r.SetEdns0(dns.DefaultMsgSize, false)
+	}
+
 	upstream := net.JoinHostPort(upstreamIP.String(), strconv.FormatUint(uint64(d.forwarderPort.Load()), 10))
 	ctx, cancel := context.WithTimeout(context.Background(), dnsTimeout)
 	defer cancel()
@@ -260,6 +268,13 @@ func (d *DnsInterceptor) ServeDNS(w dns.ResponseWriter, r *dns.Msg) {
 		return
 	}
 
+	if ede, ok := resutil.ExtractEDE(reply); ok {
+		resutil.SetMeta(w, "ede", fmt.Sprintf("%d %s", ede.InfoCode, ede.ExtraText))
+	}
+	if !hadEdns {
+		resutil.StripOPT(reply)
+	}
+
 	resutil.SetMeta(w, "peer", peerKey)
 
 	reply.Id = r.Id

client/internal/routemanager/manager.go

@@ -9,6 +9,7 @@ import (
 	"net/url"
 	"runtime"
 	"slices"
+	"strings"
 	"sync"
 	"sync/atomic"
 	"time"
@@ -332,6 +333,8 @@ func (m *DefaultManager) Stop(stateManager *statemanager.Manager) {
 		}
 	}
 
+	m.notifier.Close()
+
 	m.mux.Lock()
 	defer m.mux.Unlock()
 	m.clientRoutes = nil
@@ -700,6 +703,15 @@ func resolveURLsToIPs(urls []string) []net.IP {
 
 // updateRouteSelectorFromManagement updates the route selector based on the isSelected status from the management server
 func (m *DefaultManager) updateRouteSelectorFromManagement(clientRoutes route.HAMap) {
+	m.mirrorV6ExitPairSelections(clientRoutes)
+
+	// An explicit user "deselect all" must not be overridden by management auto-apply.
+	// Auto-applying an exit node here would call SelectRoutes, which clears the
+	// deselect-all flag and re-enables every route the user turned off.
+	if m.routeSelector.IsDeselectAll() {
+		return
+	}
+
 	exitNodeInfo := m.collectExitNodeInfo(clientRoutes)
 	if len(exitNodeInfo.allIDs) == 0 {
 		return
@@ -709,6 +721,24 @@ func (m *DefaultManager) updateRouteSelectorFromManagement(clientRoutes route.HA
 	m.logExitNodeUpdate(exitNodeInfo)
 }
 
+// mirrorV6ExitPairSelections keeps every synthesized "-v6" exit route's selection
+// consistent with its v4 base. The v4/v6 exit pair is a single toggle, so the v6
+// entry always follows the base: deselecting the v4 exit node also drops its ::/0
+// pair, and any stale (orphaned) explicit selection on the v6 entry is reset. This
+// runs before selection is read so both collectExitNodeInfo and FilterSelectedExitNodes
+// see consistent state, including pairs loaded from persisted selector state.
+func (m *DefaultManager) mirrorV6ExitPairSelections(clientRoutes route.HAMap) {
+	routesByNetID := make(map[route.NetID][]*route.Route, len(clientRoutes))
+	for haID, routes := range clientRoutes {
+		routesByNetID[haID.NetID()] = routes
+	}
+
+	for v6ID := range route.V6ExitMergeSet(routesByNetID) {
+		baseID := route.NetID(strings.TrimSuffix(string(v6ID), route.V6ExitSuffix))
+		m.routeSelector.SyncPairedSelection(baseID, v6ID)
+	}
+}
+
 type exitNodeInfo struct {
 	allIDs               []route.NetID
 	selectedByManagement []route.NetID