Merge remote-tracking branch 'origin/main' into netmap_progressive_alignment

[changing behavior!] This makes the debug bundle contain the SUPERSET of the map with nm updates and config updates

.coderabbit.yaml

@@ -1,18 +0,0 @@
-# yaml-language-server: $schema=https://coderabbit.ai/integrations/schema.v2.json
-language: en-US
-reviews:
-  profile: chill
-  request_changes_workflow: false
-  high_level_summary: true
-  poem: false
-  review_status: true
-  auto_review:
-    enabled: true
-    drafts: false
-  path_filters:
-    - "!**/*.tsx"
-    - "!**/*.ts"
-    - "!**/*.js"
-    - "!**/*.svg"
-chat:
-  auto_reply: true

.devcontainer/Dockerfile

@@ -6,6 +6,7 @@ RUN apt-get update && export DEBIAN_FRONTEND=noninteractive \
         iptables=1.8.9-2 \
         libgl1-mesa-dev=22.3.6-1+deb12u1 \
         xorg-dev=1:7.7+23 \
+        libayatana-appindicator3-dev=0.5.92-1 \
     && apt-get clean \
     && rm -rf /var/lib/apt/lists/* \
     && go install -v golang.org/x/tools/gopls@latest

.github/workflows/agent-network-e2e.yml

@@ -5,13 +5,6 @@ on:
   schedule:
     - cron: "0 3 * * *"
   workflow_dispatch:
-    inputs:
-      bedrock_model:
-        description: >-
-          Bedrock inference-profile id to drive the matrix with, exactly as
-          AWS issues it. Leave empty for the Sonnet 4.6 default.
-        required: false
-        default: ""
 
 concurrency:
   group: ${{ github.workflow }}-${{ github.ref }}
@@ -58,9 +51,6 @@ jobs:
           # token (and URL, for gateways) is unset, so partial coverage is fine.
           OPENAI_TOKEN: ${{ secrets.E2E_OPENAI_TOKEN }}
           ANTHROPIC_TOKEN: ${{ secrets.E2E_ANTHROPIC_TOKEN }}
-          # Moonshot AI platform key (platform.kimi.ai); drives both Kimi wire
-          # shapes (OpenAI /v1 and Anthropic /anthropic) through kimi_api.
-          KIMI_TOKEN: ${{ secrets.E2E_KIMI_TOKEN }}
           VERCEL_URL: ${{ secrets.E2E_VERCEL_URL }}
           VERCEL_TOKEN: ${{ secrets.E2E_VERCEL_TOKEN }}
           OPENROUTER_URL: ${{ secrets.E2E_OPENROUTER_URL }}
@@ -69,8 +59,6 @@ jobs:
           CLOUDFLARE_TOKEN: ${{ secrets.E2E_CLOUDFLARE_TOKEN }}
           AWS_BEARER_TOKEN_BEDROCK: ${{ secrets.E2E_AWS_BEARER_TOKEN_BEDROCK }}
           AWS_REGION: ${{ secrets.E2E_AWS_REGION }}
-          # Bedrock model override: dispatch input wins, then the repo variable, else the test default.
-          AWS_BEDROCK_MODEL: ${{ inputs.bedrock_model || vars.E2E_AWS_BEDROCK_MODEL }}
           # Vertex (Anthropic-on-Vertex): SA + project required; region defaults
           # to "global", model to a pinned claude snapshot.
           GOOGLE_VERTEX_SA_BASE64: ${{ secrets.E2E_GOOGLE_VERTEX_SA_BASE64 }}

.github/workflows/frontend-ui.yml

@@ -1,98 +0,0 @@
-name: UI Frontend
-
-on:
-  pull_request:
-    paths:
-      - "client/ui/frontend/**"
-      - "client/ui/i18n/**"
-      - "client/ui/**/*.go"
-      - ".github/workflows/frontend-ui.yml"
-  push:
-    branches:
-      - main
-    paths:
-      - "client/ui/frontend/**"
-      - "client/ui/i18n/**"
-      - "client/ui/**/*.go"
-
-permissions:
-  contents: read
-
-concurrency:
-  group: ${{ github.workflow }}-${{ github.ref }}-${{ github.head_ref || github.actor_id }}
-  cancel-in-progress: true
-
-jobs:
-  lint-and-build:
-    name: Lint & Build
-    runs-on: ubuntu-latest
-    timeout-minutes: 15
-    defaults:
-      run:
-        working-directory: client/ui/frontend
-    steps:
-      - name: Checkout repository
-        uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2
-        with:
-          persist-credentials: false
-
-      - name: Set up Node.js
-        uses: actions/setup-node@v4
-        with:
-          node-version: "22"
-
-      - name: Set up pnpm
-        uses: pnpm/action-setup@a3252b78c470c02df07e9d59298aecedc3ccdd6d # v3.0.0
-        with:
-          version: 11
-
-      # Bindings are generated by wails3 from the Go service definitions and
-      # are not checked in (see client/ui/frontend/bindings/). Without them,
-      # typecheck/build fail on missing module imports.
-      - name: Set up Go
-        uses: actions/setup-go@4b73464bb391d4059bd26b0524d20df3927bd417 # v6.3.0
-        with:
-          go-version-file: "go.mod"
-          cache: false
-
-      # wails3 CLI links against GTK4 / WebKitGTK 6.0 via its internal/operatingsystem
-      # package, so the dev libraries must be present before `go install`.
-      - name: Install Wails Linux system dependencies
-        run: |
-          sudo apt-get update
-          sudo apt-get install -y --no-install-recommends \
-            pkg-config \
-            libgtk-4-dev \
-            libwebkitgtk-6.0-dev
-
-      - name: Install wails3 CLI
-        # Version derived from go.mod so the binding generator always matches
-        # the wails runtime the daemon links against.
-        working-directory: ${{ github.workspace }}
-        run: |
-          WAILS_VERSION=$(go list -m -f '{{.Version}}' github.com/wailsapp/wails/v3)
-          go install github.com/wailsapp/wails/v3/cmd/wails3@$WAILS_VERSION
-
-      - name: Get pnpm store directory
-        id: pnpm-store
-        run: echo "path=$(pnpm store path --silent)" >> "$GITHUB_OUTPUT"
-
-      - name: Cache pnpm store
-        uses: actions/cache@v4
-        with:
-          path: ${{ steps.pnpm-store.outputs.path }}
-          key: ${{ runner.os }}-pnpm-${{ hashFiles('client/ui/frontend/pnpm-lock.yaml') }}
-          restore-keys: |
-            ${{ runner.os }}-pnpm-
-
-      - name: Install dependencies
-        run: pnpm install --frozen-lockfile --ignore-scripts
-
-      - name: Generate Wails bindings
-        run: pnpm run bindings
-
-      - name: Lint, typecheck, format
-        run: pnpm check
-
-      - name: Build
-        run: pnpm build

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

@@ -45,15 +45,7 @@ jobs:
         run: git --no-pager diff --exit-code
 
       - name: Test
-        # Exclude client/ui: its main.go uses //go:embed all:frontend/dist,
+        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 $(go list ./... | grep -v -e /management -e /signal -e /relay -e /proxy -e /combined -e /client/testutil/privileged)
-        # which fails to compile until the frontend has been built. The Wails UI
-        # has no Go-side unit tests, and its release pipeline runs `pnpm build`
-        # before goreleaser.
-        # `go list -e` lets the listing succeed even though the embed fails to
-        # resolve; the grep then drops the broken package by path. Without -e,
-        # go list aborts with empty stdout and `go test` falls back to the repo
-        # root, which has no Go files.
-        run: NETBIRD_STORE_ENGINE=${{ 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 -e ./... | grep -v -e /management -e /signal -e /relay -e /proxy -e /combined -e /client/ui -e /client/testutil/privileged)
 
       - name: Upload coverage reports to Codecov
         uses: codecov/codecov-action@fb8b3582c8e4def4969c97caa2f19720cb33a72f #v7.0.0

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

@@ -53,7 +53,7 @@ jobs:
 
       - name: Install dependencies
         if: steps.cache.outputs.cache-hit != 'true'
-        run: sudo apt update && sudo apt install -y -q libgtk-4-dev libwebkitgtk-6.0-dev libsoup-3.0-dev libgl1-mesa-dev xorg-dev gcc-multilib libpcap-dev
+        run: sudo apt update && sudo apt install -y -q libgtk-3-dev libayatana-appindicator3-dev libgl1-mesa-dev xorg-dev gcc-multilib libpcap-dev
 
       - name: Install 32-bit libpcap
         if: steps.cache.outputs.cache-hit != 'true'
@@ -145,7 +145,7 @@ jobs:
             ${{ runner.os }}-gotest-cache-
 
       - name: Install dependencies
-        run: sudo apt update && sudo apt install -y -q libgtk-4-dev libwebkitgtk-6.0-dev libsoup-3.0-dev libgl1-mesa-dev xorg-dev gcc-multilib libpcap-dev
+        run: sudo apt update && sudo apt install -y -q libgtk-3-dev libayatana-appindicator3-dev libgl1-mesa-dev xorg-dev gcc-multilib libpcap-dev
 
       - name: Install 32-bit libpcap
         if: matrix.arch == '386'
@@ -158,15 +158,7 @@ jobs:
         run: git --no-pager diff --exit-code
 
       - name: Test
-        # Exclude client/ui: its main.go uses //go:embed all:frontend/dist,
+        run: CGO_ENABLED=1 GOARCH=${{ matrix.arch }} CI=true go test -coverprofile=coverage.txt -tags 'devcert privileged' -exec 'sudo --preserve-env=CI,CGO_ENABLED' -timeout 10m -p 1 $(go list ./... | grep -v -e /management -e /signal -e /relay -e /proxy -e /combined -e /client/testutil/privileged)
-        # which fails to compile until the frontend has been built. The Wails UI
-        # has no Go-side unit tests, and its release pipeline runs `pnpm build`
-        # before goreleaser.
-        # `go list -e` lets the listing succeed even though the embed fails to
-        # resolve; the grep then drops the broken package by path. Without -e,
-        # go list aborts with empty stdout and `go test` falls back to the repo
-        # root, which has no Go files.
-        run: CGO_ENABLED=1 GOARCH=${{ matrix.arch }} CI=true go test -coverprofile=coverage.txt -tags 'devcert privileged' -exec 'sudo --preserve-env=CI,CGO_ENABLED' -timeout 10m -p 1 $(go list -e ./... | grep -v -e /management -e /signal -e /relay -e /proxy -e /combined -e /client/ui -e /client/testutil/privileged)
 
       - name: Upload coverage reports to Codecov
         if: matrix.arch == 'amd64'
@@ -176,6 +168,7 @@ jobs:
           slug: netbirdio/netbird
           flags: unit,client
 
+
   test_client_on_docker:
     name: "Client (Docker) / Unit"
     needs: [build-cache]
@@ -236,7 +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 privileged" -v -timeout 10m -p 1 $(go list -e -buildvcs=false ./... | grep -v -e /management -e /signal -e /relay -e /proxy -e /combined -e /client/ui -e /upload-server -e /client/testutil/privileged)
+              go test -buildvcs=false -tags "devcert privileged" -v -timeout 10m -p 1 $(go list -buildvcs=false ./... | grep -v -e /management -e /signal -e /relay -e /proxy -e /combined -e /client/ui -e /upload-server -e /client/testutil/privileged)
             '
 
   test_relay:

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

@@ -65,15 +65,8 @@ jobs:
       - run: PsExec64 -s -w ${{ github.workspace }} C:\hostedtoolcache\windows\go\${{ steps.go.outputs.go-version }}\x64\bin\go.exe env -w GOCACHE=${{ env.modcache }}
       - run: PsExec64 -s -w ${{ github.workspace }} C:\hostedtoolcache\windows\go\${{ steps.go.outputs.go-version }}\x64\bin\go.exe mod tidy
       - name: Generate test script
-        # Exclude client/ui: its main.go uses //go:embed all:frontend/dist,
-        # which fails to compile until the frontend has been built. The Wails UI
-        # has no Go-side unit tests, and its release pipeline runs `pnpm build`
-        # before goreleaser.
-        # `go list -e` lets the listing succeed even though the embed fails to
-        # resolve; the Where-Object pipeline then drops the broken package by
-        # path. Without -e, go list aborts with empty stdout.
         run: |
-          $packages = go list -e ./... | Where-Object { $_ -notmatch '/management' } | Where-Object { $_ -notmatch '/relay' } | Where-Object { $_ -notmatch '/signal' } | Where-Object { $_ -notmatch '/proxy' } | Where-Object { $_ -notmatch '/combined' } | Where-Object { $_ -notmatch '/client/ui' }
+          $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 privileged`" -timeout 10m -p 1 $($packages -join ' ') > test-out.txt 2>&1"
           Set-Content -Path "${{ github.workspace }}\run-tests.cmd" -Value $cmd

.github/workflows/golangci-lint.yml

@@ -22,15 +22,7 @@ jobs:
         uses: codespell-project/actions-codespell@8f01853be192eb0f849a5c7d721450e7a467c579 # v2.2
         with:
           ignore_words_list: erro,clienta,hastable,iif,groupd,testin,groupe,cros,ans,deriver,te,userA,ede,additionals,flate,recordin,unparseable
-          # Non-English UI translations trip codespell on real foreign words
+          skip: go.mod,go.sum,**/proxy/web/**
-          # (de: "Sie", "oder", "ist"). Only en/common.json is the source of
-          # truth that should be spell-checked. List each translated locale
-          # dir below and add new ones as languages are added under
-          # client/ui/i18n/locales/. Single-star globs are matched per path
-          # segment by codespell and behave the same across versions; the
-          # recursive "**" form did not take effect with the codespell shipped
-          # by this action.
-          skip: go.mod,go.sum,*/proxy/web/*,*pnpm-lock.yaml,*package-lock.json,*/locales/de/*,*/locales/es/*,*/locales/fr/*,*/locales/hu/*,*/locales/it/*,*/locales/pt/*,*/locales/ru/*,*/locales/zh-CN/*,*/i18n/TRANSLATING.md
   golangci:
     strategy:
       fail-fast: false
@@ -45,7 +37,7 @@ jobs:
             display_name: Linux
     name: ${{ matrix.display_name }}
     runs-on: ${{ matrix.os }}
-    timeout-minutes: 25
+    timeout-minutes: 15
     steps:
       - name: Checkout code
         uses: actions/checkout@9c091bb21b7c1c1d1991bb908d89e4e9dddfe3e0 # v7.0.0
@@ -62,16 +54,7 @@ jobs:
           cache: false
       - name: Install dependencies
         if: matrix.os == 'ubuntu-latest'
-        run: sudo apt update && sudo apt install -y -q libgtk-4-dev libwebkitgtk-6.0-dev libsoup-3.0-dev libgl1-mesa-dev xorg-dev libpcap-dev
+        run: sudo apt update && sudo apt install -y -q libgtk-3-dev libayatana-appindicator3-dev libgl1-mesa-dev xorg-dev libpcap-dev
-      - name: Stub Wails frontend bundle
-        # client/ui/main.go has //go:embed all:frontend/dist. The
-        # directory is produced by `pnpm run build` and is gitignored, so
-        # lint-only runs (no frontend toolchain) need a placeholder file
-        # for the embed pattern to match.
-        shell: bash
-        run: |
-          mkdir -p client/ui/frontend/dist
-          touch client/ui/frontend/dist/.embed-placeholder
       - name: golangci-lint
         uses: golangci/golangci-lint-action@82606bf257cbaff209d206a39f5134f0cfbfd2ee #v9.2.1
         with:
@@ -79,4 +62,4 @@ jobs:
           skip-cache: true
           skip-save-cache: true
           cache-invalidation-interval: 0
-          args: --timeout=20m
+          args: --timeout=12m

.github/workflows/release.yml

@@ -9,7 +9,7 @@ on:
   pull_request:
 
 env:
-  SIGN_PIPE_VER: "v0.1.8"
+  SIGN_PIPE_VER: "v0.1.6"
   GORELEASER_VER: "v2.16.0"
   PRODUCT_NAME: "NetBird"
   COPYRIGHT: "NetBird GmbH"
@@ -216,9 +216,9 @@ jobs:
       - name: Install goversioninfo
         run: go install github.com/josephspurrier/goversioninfo/cmd/goversioninfo@233067e
       - name: Generate windows syso amd64
-        run: goversioninfo  -icon client/ui/build/windows/icon.ico -manifest client/manifest.xml -product-name ${{ env.PRODUCT_NAME }} -copyright "${{ env.COPYRIGHT }}" -ver-major ${{ steps.semver_parser.outputs.major }} -ver-minor ${{ steps.semver_parser.outputs.minor }} -ver-patch ${{ steps.semver_parser.outputs.patch }} -ver-build 0 -file-version ${{ steps.semver_parser.outputs.fullversion }}.0 -product-version ${{ steps.semver_parser.outputs.fullversion }}.0 -o client/resources_windows_amd64.syso
+        run: goversioninfo  -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_amd64.syso
       - name: Generate windows syso arm64
-        run: goversioninfo -arm -64 -icon client/ui/build/windows/icon.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
+        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@5daf1e915a5f0af01ddbcd89a43b8061ff4f1a89 # v7.2.2
@@ -397,18 +397,8 @@ jobs:
       - name: check git status
         run: git --no-pager diff --exit-code
 
-      - name: Set up Node.js
-        uses: actions/setup-node@v4
-        with:
-          node-version: '22'
-
-      - name: Set up pnpm
-        uses: pnpm/action-setup@a3252b78c470c02df07e9d59298aecedc3ccdd6d # v3.0.0
-        with:
-          version: 11
-
       - name: Install dependencies
-        run: sudo apt update && sudo apt install -y -q libgtk-4-dev libwebkitgtk-6.0-dev libsoup-3.0-dev gcc-mingw-w64-x86-64
+        run: sudo apt update && sudo apt install -y -q libappindicator3-dev gir1.2-appindicator3-0.1 libxxf86vm-dev gcc-mingw-w64-x86-64
 
       - name: Decode GPG signing key
         if: github.event_name != 'pull_request' || github.event.pull_request.head.repo.full_name == github.repository
@@ -427,16 +417,10 @@ jobs:
           echo "/tmp/llvm-mingw-20250709-ucrt-ubuntu-22.04-x86_64/bin" >> $GITHUB_PATH
       - name: Install goversioninfo
         run: go install github.com/josephspurrier/goversioninfo/cmd/goversioninfo@233067e
-      - name: Install wails3 CLI
-        # Version derived from go.mod so the binding generator always matches
-        # the wails runtime the binary links against.
-        run: |
-          WAILS_VERSION=$(go list -m -f '{{.Version}}' github.com/wailsapp/wails/v3)
-          go install github.com/wailsapp/wails/v3/cmd/wails3@$WAILS_VERSION
       - name: Generate windows syso amd64
-        run: goversioninfo -64 -icon client/ui/build/windows/icon.ico -manifest client/ui/build/windows/wails.exe.manifest -product-name ${{ env.PRODUCT_NAME }}-"UI" -copyright "${{ env.COPYRIGHT }}" -ver-major ${{ steps.semver_parser.outputs.major }} -ver-minor ${{ steps.semver_parser.outputs.minor }} -ver-patch ${{ steps.semver_parser.outputs.patch }} -ver-build 0 -file-version ${{ steps.semver_parser.outputs.fullversion }}.0 -product-version ${{ steps.semver_parser.outputs.fullversion }}.0 -o client/ui/resources_windows_amd64.syso
+        run: goversioninfo -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_amd64.syso
       - name: Generate windows syso arm64
-        run: goversioninfo -arm -64 -icon client/ui/build/windows/icon.ico -manifest client/ui/build/windows/wails.exe.manifest -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
+        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@5daf1e915a5f0af01ddbcd89a43b8061ff4f1a89 # v7.2.2
@@ -505,20 +489,6 @@ jobs:
         run: go mod tidy
       - name: check git status
         run: git --no-pager diff --exit-code
-      - name: Set up Node.js
-        uses: actions/setup-node@v4
-        with:
-          node-version: '22'
-      - name: Set up pnpm
-        uses: pnpm/action-setup@a3252b78c470c02df07e9d59298aecedc3ccdd6d # v3.0.0
-        with:
-          version: 11
-      - name: Install wails3 CLI
-        # Version derived from go.mod so the binding generator always matches
-        # the wails runtime the binary links against.
-        run: |
-          WAILS_VERSION=$(go list -m -f '{{.Version}}' github.com/wailsapp/wails/v3)
-          go install github.com/wailsapp/wails/v3/cmd/wails3@$WAILS_VERSION
       - name: Run GoReleaser
         id: goreleaser
         uses: goreleaser/goreleaser-action@5daf1e915a5f0af01ddbcd89a43b8061ff4f1a89 # v7.2.2
@@ -606,6 +576,23 @@ jobs:
       - name: Move wintun.dll into dist
         run: mv ${{ env.downloadPath }}\wintun\bin\${{ matrix.wintun_arch }}\wintun.dll ${{ github.workspace }}\dist\${{ env.PackageWorkdir }}\
 
+      - name: Download Mesa3D (amd64 only)
+        id: download-mesa3d
+        if: matrix.arch == 'amd64'
+        uses: netbirdio/shared-actions/actions/win-download-and-verify@be5df6047383da2236e02243cceb857d8567c27e # v0.0.2
+        with:
+          url: https://pkgs.netbird.io/mesa3d/MesaForWindows-x64-20.1.8.7z
+          destination: ${{ env.downloadPath }}\mesa3d.7z
+          sha256: 71c7cb64ec229a1d6b8d62fa08e1889ed2bd17c0eeede8689daf0f25cb31d6b9
+
+      - name: Extract Mesa3D driver (amd64 only)
+        if: matrix.arch == 'amd64'
+        run: 7z x -o"${{ env.downloadPath }}" "${{ env.downloadPath }}/mesa3d.7z"
+
+      - name: Move opengl32.dll into dist (amd64 only)
+        if: matrix.arch == 'amd64'
+        run: mv ${{ env.downloadPath }}\opengl32.dll ${{ github.workspace }}\dist\${{ env.PackageWorkdir }}\
+
       - name: Download EnVar plugin for NSIS
         uses: netbirdio/shared-actions/actions/win-download-and-verify@be5df6047383da2236e02243cceb857d8567c27e # v0.0.2
         with:
@@ -628,28 +615,6 @@ jobs:
         if: matrix.arch == 'amd64'
         run: 7z x -o"${{ github.workspace }}/NSIS_Plugins" "${{ github.workspace }}/ShellExecAsUser_amd64-Unicode.7z"
 
-      - name: Set up Go for wails3 CLI
-        uses: actions/setup-go@v5
-        with:
-          go-version-file: "go.mod"
-          cache: false
-
-      - name: Install wails3 CLI
-        # Version derived from go.mod so the bootstrapper payload always
-        # matches the wails runtime the binary links against.
-        shell: bash
-        run: |
-          WAILS_VERSION=$(go list -m -f '{{.Version}}' github.com/wailsapp/wails/v3)
-          go install github.com/wailsapp/wails/v3/cmd/wails3@$WAILS_VERSION
-
-      - name: Stage WebView2 bootstrapper for installers
-        # Both client/installer.nsis and client/netbird.wxs reference
-        # client/MicrosoftEdgeWebview2Setup.exe. wails3 writes it there.
-        # The signing pipeline (netbirdio/sign-pipelines) does the same
-        # step for release builds; this mirrors it for PR sanity testing.
-        shell: bash
-        run: wails3 generate webview2bootstrapper -dir client
-
       - name: Build NSIS installer
         shell: pwsh
         env:

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

@@ -27,7 +27,7 @@ jobs:
         with:
           go-version-file: "go.mod"
       - name: Install dependencies
-        run: sudo apt update && sudo apt install -y -q libgtk-4-dev libwebkitgtk-6.0-dev libsoup-3.0-dev libgl1-mesa-dev xorg-dev libpcap-dev
+        run: sudo apt update && sudo apt install -y -q libgtk-3-dev libayatana-appindicator3-dev libgl1-mesa-dev xorg-dev libpcap-dev
       - name: Install golangci-lint
         uses: golangci/golangci-lint-action@82606bf257cbaff209d206a39f5134f0cfbfd2ee #v9.2.1
         with:

.golangci.yaml

@@ -114,16 +114,6 @@ linters:
       - linters:
           - staticcheck
         text: "QF1012"
-      # client/ui/main.go uses //go:embed all:frontend/dist; the
-      # directory is populated by `pnpm build` in the release pipeline
-      # and missing at lint time, so the embed parses to "no matching
-      # files found" — surfaced by golangci-lint's typecheck pre-pass.
-      # Suppress just that one diagnostic; the rest of the package
-      # (services/, tray.go, grpc.go, ...) still gets linted normally.
-      - linters:
-          - typecheck
-        path: client/ui/main\.go
-        text: "pattern all:frontend/dist"
     paths:
       - third_party$
       - builtin$

.goreleaser.yaml

@@ -212,7 +212,6 @@ nfpms:
     description: Netbird client.
     homepage: https://netbird.io/
     license: BSD-3-Clause
-    vendor: NetBird
     id: netbird_deb
     bindir: /usr/bin
     builds:
@@ -227,7 +226,6 @@ nfpms:
     description: Netbird client.
     homepage: https://netbird.io/
     license: BSD-3-Clause
-    vendor: NetBird
     id: netbird_rpm
     bindir: /usr/bin
     builds:
@@ -273,8 +271,8 @@ dockers_v2:
        - netbirdio/netbird
        - ghcr.io/netbirdio/netbird
      tags:
-       - "{{ .Version }}-rootless"
+       - "v{{ .Version }}-rootless"
-       - "{{ if eq .Env.SKIP_PUBLISH \"false\" }}rootless-latest{{ end }}"
+       - "{{ if eq .Env.SKIP_PUBLISH \"false\" }}latest{{ end }}"
      dockerfile: client/Dockerfile-rootless
      extra_files:
        - client/netbird-entrypoint.sh

.goreleaser_ui.yaml

@@ -2,15 +2,6 @@ version: 2
 env:
   - SKIP_PUBLISH={{ if index .Env "SKIP_PUBLISH" }}{{ .Env.SKIP_PUBLISH }}{{ else }}true{{ end }}
 project_name: netbird-ui
-
-before:
-  hooks:
-    # Bindings are gitignored; regenerate before the frontend build so
-    # the @wailsio/runtime Vite plugin can resolve them (vite refuses to
-    # build without them).
-    - sh -c 'cd client/ui && wails3 generate bindings -clean=true -ts'
-    - sh -c 'cd client/ui/frontend && pnpm install --frozen-lockfile && pnpm build'
-
 builds:
   - id: netbird-ui
     dir: client/ui
@@ -71,8 +62,6 @@ nfpms:
   - maintainer: Netbird <dev@netbird.io>
     description: Netbird client UI.
     homepage: https://netbird.io/
-    license: BSD-3-Clause
-    vendor: NetBird
     id: netbird_ui_deb
     package_name: netbird-ui
     builds:
@@ -82,9 +71,9 @@ nfpms:
     scripts:
       postinstall: "release_files/ui-post-install.sh"
     contents:
-      - src: client/ui/build/linux/netbird.desktop
+      - src: client/ui/build/netbird.desktop
-        dst: /usr/share/applications/org.wails.netbird.desktop
+        dst: /usr/share/applications/netbird.desktop
-      - src: client/ui/build/appicon.png
+      - src: client/ui/assets/netbird.png
         dst: /usr/share/pixmaps/netbird.png
     dependencies:
       - netbird
@@ -92,8 +81,6 @@ nfpms:
   - maintainer: Netbird <dev@netbird.io>
     description: Netbird client UI.
     homepage: https://netbird.io/
-    license: BSD-3-Clause
-    vendor: NetBird
     id: netbird_ui_rpm
     package_name: netbird-ui
     builds:
@@ -103,13 +90,12 @@ nfpms:
     scripts:
       postinstall: "release_files/ui-post-install.sh"
     contents:
-      - src: client/ui/build/linux/netbird.desktop
+      - src: client/ui/build/netbird.desktop
-        dst: /usr/share/applications/org.wails.netbird.desktop
+        dst: /usr/share/applications/netbird.desktop
-      - src: client/ui/build/appicon.png
+      - src: client/ui/assets/netbird.png
         dst: /usr/share/pixmaps/netbird.png
     dependencies:
       - netbird
-
     rpm:
       signature:
         key_file: '{{ if index .Env "GPG_RPM_KEY_FILE" }}{{ .Env.GPG_RPM_KEY_FILE }}{{ end }}'

.goreleaser_ui_darwin.yaml

@@ -1,15 +1,6 @@
 version: 2
 
 project_name: netbird-ui
-
-before:
-  hooks:
-    # Bindings are gitignored; regenerate before the frontend build so
-    # the @wailsio/runtime Vite plugin can resolve them (vite refuses to
-    # build without them).
-    - sh -c 'cd client/ui && wails3 generate bindings -clean=true -ts'
-    - sh -c 'cd client/ui/frontend && pnpm install --frozen-lockfile && pnpm build'
-
 builds:
   - id: netbird-ui-darwin
     dir: client/ui
@@ -29,6 +20,8 @@ builds:
     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 }}"
+    tags:
+      - load_wgnt_from_rsrc
 
 universal_binaries:
   - id: netbird-ui-darwin

CONTRIBUTING.md

@@ -79,21 +79,13 @@ dependencies are installed. Here is a short guide on how that can be done.
 
 ### Requirements
 
-#### Go 1.25
+#### Go 1.21
 
 Follow the installation guide from https://go.dev/
 
-#### UI client - Wails v3 + React
+#### UI client - Fyne toolkit 
 
-The desktop UI client (`client/ui`) is built with [Wails v3](https://v3.wails.io/) and a React frontend rendered in a WebView. To build it you need:
+We use the fyne toolkit in our UI client. You can follow its requirement guide to have all its dependencies installed: https://developer.fyne.io/started/#prerequisites
-
-- Go ≥ 1.25
-- Node ≥ 20 and **pnpm** (`corepack enable && corepack prepare pnpm@latest --activate`)
-- The `wails3` CLI: `go install github.com/wailsapp/wails/v3/cmd/wails3@latest`
-- The `task` runner: `go install github.com/go-task/task/v3/cmd/task@latest`
-- Linux only: `libwebkitgtk-6.0-dev`, `libgtk-4-dev`, `libsoup-3.0-dev`
-
-All UI build, dev-loop, and cross-compile commands are described in the [UI client](#ui-client) section below.
 
 #### gRPC
 You can follow the instructions from the quickstarter guide https://grpc.io/docs/languages/go/quickstart/#prerequisites and then run the `generate.sh` files located in each `proto` directory to generate changes.
@@ -222,39 +214,6 @@ To start NetBird the client in the foreground:
 sudo ./client up --log-level debug --log-file console
 ```
 > On Windows use a powershell with administrator privileges
-
-#### UI client
-
-The desktop UI lives in `client/ui` and is built with Wails v3 (see [Requirements](#ui-client---wails-v3--react)). All commands run from `client/ui`.
-
-Live-reload development (Vite + Go binary + `*.go` watcher):
-
-```
-cd client/ui
-task dev
-```
-
-Pass daemon flags after `--`:
-
-```
-task dev -- --daemon-addr=tcp://127.0.0.1:41731
-```
-
-Production build (frontend assets embedded into the binary, output in `client/ui/bin/`):
-
-```
-cd client/ui
-task build
-```
-
-Cross-compile the Windows binary from Linux (requires the mingw-w64 toolchain, e.g. `sudo apt install gcc-mingw-w64-x86-64`):
-
-```
-CGO_ENABLED=1 task windows:build
-```
-
-> macOS cross-compile from Linux is not supported (signing and notarization need a real Mac).
-
 #### Signal service
 
 To start NetBird's signal, execute:
@@ -292,10 +251,10 @@ Create dist directory
 mkdir -p dist/netbird_windows_amd64
 ```
 
-UI client (built with Wails v3 — see the [UI client](#ui-client) section above)
+UI client
 ```shell
-(cd client/ui && CGO_ENABLED=1 task windows:build)
+CC=x86_64-w64-mingw32-gcc CGO_ENABLED=1 GOOS=windows GOARCH=amd64 go build -o netbird-ui.exe -ldflags "-s -w -H windowsgui" ./client/ui
-mv client/ui/bin/netbird-ui.exe ./dist/netbird_windows_amd64/
+mv netbird-ui.exe ./dist/netbird_windows_amd64/
 ```
 
 Client
@@ -332,6 +291,8 @@ go test -exec sudo ./...
 ```
 > On Windows use a powershell with administrator privileges
 
+> Non-GTK environments will need the `libayatana-appindicator3-dev` (debian/ubuntu) package installed
+
 ## Checklist before submitting a PR
 As a critical network service and open-source project, we must enforce a few things before submitting the pull-requests:
 - Keep functions as simple as possible, with a single purpose

agent-network/README.md

@@ -1,47 +1,16 @@
 # NetBird Agent Network
 
-Agent Network is NetBird's access control layer for AI agents and the people who run them. 
+Agent Network is NetBird's access control layer for AI agents and the people who run
-It gives every agent a real identity, tied to an identity provider (IdP), and governs what it can reach: LLM APIs and 
+them. It gives every agent a real identity, tied to your identity provider (IdP), and
-AI gateways it can call, and the internal resources it can access. Traffic flows only over the encrypted NetBird tunnel, 
+governs what it can reach — the LLM APIs and AI gateways it can call, and the internal
-scoped by policy, with no API keys or other credentials to leak. It also gives you control over cost and token usage.
+resources it can access. Traffic flows only over the encrypted NetBird tunnel, scoped by
+policy, with no API keys to leak.
 
-Because every LLM request passes through an
+> **Beta.** Agent Network is open source and can be self-hosted on your own
-identity-aware proxy, you can:
+> infrastructure.
-
-- **Set spending and rate limits** per agent, per user, or per team — with hard caps
-  that stop requests once a budget is reached.
-- **Restrict models and providers** so agents can only call approved (and cost-appropriate)
-  endpoints, keeping expensive models off-limits unless explicitly allowed.
-- **Attribute usage** by tracking token consumption and cost per identity, group, or cost center so every
-  request is tied back to the agent and person responsible.
-- **Reuse your existing AI gateway** — point the proxy at a gateway you already run,
-  keeping its routing and config in place while it adds identity on top, so you skip
-  API key distribution.
-
-https://github.com/user-attachments/assets/44d18286-d8ab-49f8-a457-98ccd66f3268
-
-> **Beta.** Agent Network is in beta, but it's stable and already running in
-> production environments. It's fully open source and can be self-hosted on your own
-> infrastructure, with no vendor lock-in and no data leaving your environment.
 
 ## How it works
 
-Say you have a simple use case: your Engineering or IT team needs access to Claude Code or Codex, and you want visibility into usage plus the ability to enforce budgets.
-How can you do that without creating a dedicated API key for every team?
-
-With Agent Network you get a private endpoint inside your network, for example: https://mirror.netbird.ai
-Teams configure their agents to point to that endpoint instead of using individual API keys directly. 
-
-This endpoint is only reachable when users are connected to your NetBird network and authenticated through your IdP. Otherwise, it is not accessible from the public internet.
-You can then use this private endpoint to configure your AI agents, whether that is Claude Code, Codex, or another tool.
-
-## Quickstart
-
-Full step-by-step setup:
-**https://docs.netbird.io/agent-network/quickstart**
-
-## Architecture 
-
 Agent Network is built on two existing NetBird capabilities:
 
 - **Overlay network** — the encrypted WireGuard mesh between peers.
@@ -53,9 +22,6 @@ LLM traffic is routed through the proxy's identity-aware pipeline, while interna
 resources (databases, internal APIs, self-hosted models) are reached directly over
 peer-to-peer WireGuard tunnels, governed by the same identities and access policies.
 
-<img width="4720" height="2218" alt="image" src="https://github.com/user-attachments/assets/1afa5da1-4b82-4f8a-a7a8-f417efadf1eb" />
-
-
 ## Where the code lives
 
 There is no separate "agent-network" service — it reuses the reverse-proxy and management

client/android/client.go

@@ -7,7 +7,6 @@ import (
 	"fmt"
 	"os"
 	"slices"
-	"strings"
 	"sync"
 	"time"
 
@@ -76,24 +75,6 @@ type Client struct {
 	connectClient *internal.ConnectClient
 	config        *profilemanager.Config
 	cacheDir      string
-
-	stateChangeMu    sync.Mutex
-	stateChangeSubID string
-	eventSub         *peer.EventSubscription
-	// Closed to stop the watch goroutines from delivering buffered items to a
-	// listener that has been removed or replaced. See stopStateChangeWatchLocked.
-	stateChangeDone chan struct{}
-
-	// Latched "the server wants an interactive login": survives the engine
-	// restarts that replace the run loop's context state. See Client.Status.
-	// Guarded by loginRequiredMu together with loginCleared, which counts
-	// clears so a stale observation cannot re-latch over one.
-	loginRequiredMu sync.Mutex
-	loginRequired   bool
-	loginCleared    uint64
-
-	extendMu     sync.Mutex
-	extendCancel context.CancelFunc
 }
 
 func (c *Client) setState(cfg *profilemanager.Config, cacheDir string, cc *internal.ConnectClient) {
@@ -167,16 +148,11 @@ func (c *Client) Run(platformFiles PlatformFiles, urlOpener URLOpener, isAndroid
 	if err != nil {
 		return err
 	}
+
 	// todo do not throw error in case of cancelled context
 	ctx = internal.CtxInitState(ctx)
 	connectClient := internal.NewConnectClient(ctx, cfg, c.recorder)
 	c.setState(cfg, cacheDir, connectClient)
-	// This path runs the interactive SSO flow, so reaching here means the peer
-	// is authenticated again — release the latch Status() reports from. Clear
-	// only once the fresh connect client is installed: until then Status()
-	// still reads the previous run's context state, which holds the NeedsLogin
-	// that prompted this login, and would re-latch what was just cleared.
-	c.clearLoginRequired()
 	return connectClient.RunOnAndroid(c.tunAdapter, c.iFaceDiscover, c.networkChangeListener, slices.Clone(dns.items), dnsReadyListener, stateFile, cacheDir)
 }
 
@@ -271,9 +247,6 @@ func (c *Client) DebugBundle(platformFiles PlatformFiles, anonymize bool) (strin
 		deps.SyncResponse = resp
 
 		if e := cc.Engine(); e != nil {
-			deps.RefreshStatus = func() {
-				e.RunHealthProbes(context.Background(), true)
-			}
 			if cm := e.GetClientMetrics(); cm != nil {
 				deps.ClientMetrics = cm
 			}
@@ -323,13 +296,6 @@ func (c *Client) SetInfoLogLevel() {
 // PeersList return with the list of the PeerInfos
 func (c *Client) PeersList() *PeerInfoArray {
 
-	// The recorder only caches transfer counters and handshake times; nothing
-	// refreshes them on its own, so without this they read as zero. The desktop
-	// daemon does the same before serving a full peer status.
-	if err := c.recorder.RefreshWireGuardStats(); err != nil {
-		log.Debugf("failed to refresh WireGuard stats: %v", err)
-	}
-
 	fullStatus := c.recorder.GetFullStatus()
 
 	peerInfos := make([]PeerInfo, len(fullStatus.Peers))
@@ -340,20 +306,6 @@ func (c *Client) PeersList() *PeerInfoArray {
 			FQDN:       p.FQDN,
 			ConnStatus: int(p.ConnStatus),
 			Routes:     PeerRoutes{routes: maps.Keys(p.GetRoutes())},
-
-			PubKey:                     p.PubKey,
-			Latency:                    formatDuration(p.Latency),
-			LatencyMs:                  p.Latency.Milliseconds(),
-			BytesRx:                    p.BytesRx,
-			BytesTx:                    p.BytesTx,
-			ConnStatusUpdate:           formatTime(p.ConnStatusUpdate),
-			Relayed:                    p.Relayed,
-			RosenpassEnabled:           p.RosenpassEnabled,
-			LastWireguardHandshake:     formatTime(p.LastWireguardHandshake),
-			LocalIceCandidateType:      p.LocalIceCandidateType,
-			RemoteIceCandidateType:     p.RemoteIceCandidateType,
-			LocalIceCandidateEndpoint:  p.LocalIceCandidateEndpoint,
-			RemoteIceCandidateEndpoint: p.RemoteIceCandidateEndpoint,
 		}
 		peerInfos[n] = pi
 	}
@@ -484,6 +436,10 @@ func (c *Client) RemoveConnectionListener() {
 	c.recorder.RemoveConnectionListener()
 }
 
+func (c *Client) toggleRoute(command routeCommand) error {
+	return command.toggleRoute()
+}
+
 func (c *Client) getRouteManager() (routemanager.Manager, error) {
 	client := c.getConnectClient()
 	if client == nil {
@@ -503,22 +459,22 @@ func (c *Client) getRouteManager() (routemanager.Manager, error) {
 	return manager, nil
 }
 
-func (c *Client) SelectRoute(id string) error {
+func (c *Client) SelectRoute(route string) error {
 	manager, err := c.getRouteManager()
 	if err != nil {
 		return err
 	}
 
-	return manager.SelectRoutes([]route.NetID{route.NetID(id)}, true)
+	return c.toggleRoute(selectRouteCommand{route: route, manager: manager})
 }
 
-func (c *Client) DeselectRoute(id string) error {
+func (c *Client) DeselectRoute(route string) error {
 	manager, err := c.getRouteManager()
 	if err != nil {
 		return err
 	}
 
-	return manager.DeselectRoutes([]route.NetID{route.NetID(id)})
+	return c.toggleRoute(deselectRouteCommand{route: route, manager: manager})
 }
 
 // getNetworkDomainsFromRoute extracts domains from a route and enriches each domain
@@ -553,28 +509,3 @@ func exportEnvList(list *EnvList) {
 		}
 	}
 }
-
-// formatDuration renders a duration for display, trimming the fractional part
-// to two digits so latencies read as "12.34ms" rather than "12.345678ms".
-func formatDuration(d time.Duration) string {
-	ds := d.String()
-	dotIndex := strings.Index(ds, ".")
-	if dotIndex == -1 {
-		return ds
-	}
-
-	endIndex := min(dotIndex+3, len(ds))
-
-	// Skip the remaining digits so only the unit suffix is appended back.
-	unitStart := endIndex
-	for unitStart < len(ds) && ds[unitStart] >= '0' && ds[unitStart] <= '9' {
-		unitStart++
-	}
-	return ds[:endIndex] + ds[unitStart:]
-}
-
-// formatTime renders a timestamp in UTC using a fixed layout. The zero time is
-// passed through as-is so the UI can recognise it and show "never" instead.
-func formatTime(t time.Time) string {
-	return t.UTC().Format("2006-01-02 15:04:05")
-}

client/android/peer_notifier.go

@@ -12,30 +12,12 @@ const (
 )
 
 // PeerInfo describe information about the peers. It designed for the UI usage
-//
-// The fields below ConnStatus back the peer detail screen. Durations and times
-// are pre-formatted into strings so the UI does not have to know Go's layouts;
-// Latency is additionally exposed as LatencyMs for colour coding.
 type PeerInfo struct {
 	IP         string
 	IPv6       string
 	FQDN       string
 	ConnStatus int
 	Routes     PeerRoutes
-
-	PubKey                     string
-	Latency                    string
-	LatencyMs                  int64
-	BytesRx                    int64
-	BytesTx                    int64
-	ConnStatusUpdate           string
-	Relayed                    bool
-	RosenpassEnabled           bool
-	LastWireguardHandshake     string
-	LocalIceCandidateType      string
-	RemoteIceCandidateType     string
-	LocalIceCandidateEndpoint  string
-	RemoteIceCandidateEndpoint string
 }
 
 func (p *PeerInfo) GetPeerRoutes() *PeerRoutes {

client/android/profile_manager.go

@@ -189,19 +189,6 @@ func (pm *ProfileManager) LogoutProfile(id string) error {
 	return nil
 }
 
-// RenameProfile changes a profile's display name. The profile ID, and therefore
-// its on-disk filename, is left untouched: only the "name" field of the config
-// is rewritten. This works for the default profile too, whose config lives in
-// netbird.cfg rather than under profiles/.
-func (pm *ProfileManager) RenameProfile(id string, newName string) error {
-	if err := pm.serviceMgr.RenameProfile(profilemanager.ID(id), androidUsername, newName); err != nil {
-		return fmt.Errorf("failed to rename profile: %w", err)
-	}
-
-	log.Infof("renamed profile %s to: %s", id, newName)
-	return nil
-}
-
 // RemoveProfile deletes a profile
 func (pm *ProfileManager) RemoveProfile(id string) error {
 	// Use ServiceManager (removes profile from profiles/ directory)

client/android/route_command.go

@@ -0,0 +1,70 @@
+//go:build android
+
+package android
+
+import (
+	"fmt"
+
+	log "github.com/sirupsen/logrus"
+	"golang.org/x/exp/maps"
+
+	"github.com/netbirdio/netbird/client/internal/routemanager"
+	"github.com/netbirdio/netbird/route"
+)
+
+func executeRouteToggle(id string, manager routemanager.Manager,
+	operationName string,
+	routeOperation func(routes []route.NetID, allRoutes []route.NetID) error) error {
+	netID := route.NetID(id)
+	routes := []route.NetID{netID}
+
+	routesMap := manager.GetClientRoutesWithNetID()
+	routes = route.ExpandV6ExitPairs(routes, routesMap)
+
+	log.Debugf("%s with ids: %v", operationName, routes)
+
+	if err := routeOperation(routes, maps.Keys(routesMap)); err != nil {
+		log.Debugf("error when %s: %s", operationName, err)
+		return fmt.Errorf("error %s: %w", operationName, err)
+	}
+
+	manager.TriggerSelection(manager.GetClientRoutes())
+
+	return nil
+}
+
+type routeCommand interface {
+	toggleRoute() error
+}
+
+type selectRouteCommand struct {
+	route   string
+	manager routemanager.Manager
+}
+
+func (s selectRouteCommand) toggleRoute() error {
+	routeSelector := s.manager.GetRouteSelector()
+	if routeSelector == nil {
+		return fmt.Errorf("no route selector available")
+	}
+
+	routeOperation := func(routes []route.NetID, allRoutes []route.NetID) error {
+		return routeSelector.SelectRoutes(routes, true, allRoutes)
+	}
+
+	return executeRouteToggle(s.route, s.manager, "selecting route", routeOperation)
+}
+
+type deselectRouteCommand struct {
+	route   string
+	manager routemanager.Manager
+}
+
+func (d deselectRouteCommand) toggleRoute() error {
+	routeSelector := d.manager.GetRouteSelector()
+	if routeSelector == nil {
+		return fmt.Errorf("no route selector available")
+	}
+
+	return executeRouteToggle(d.route, d.manager, "deselecting route", routeSelector.DeselectRoutes)
+}

client/android/session.go

@@ -1,309 +0,0 @@
-//go:build android
-
-package android
-
-import (
-	"context"
-	"fmt"
-
-	log "github.com/sirupsen/logrus"
-
-	"github.com/netbirdio/netbird/client/internal"
-	"github.com/netbirdio/netbird/client/internal/auth"
-	"github.com/netbirdio/netbird/client/internal/auth/sessionwatch"
-	"github.com/netbirdio/netbird/client/internal/peer"
-	cProto "github.com/netbirdio/netbird/client/proto"
-)
-
-// StateChangeListener receives client state notifications.
-//
-// OnStateChanged is a payload-free wake-up whenever the state snapshot
-// changed: connection state, the run-loop status label (e.g. NeedsLogin) or
-// the session deadline. It mirrors the daemon's SubscribeStatus stream
-// trigger — on each signal the consumer pulls the fresh values via
-// Status() / SessionExpiresAtUnix().
-//
-// OnSessionExpiring forwards the engine's session-expiry warnings, fired at
-// sessionwatch.WarningLead before the deadline and again at FinalWarningLead
-// (finalWarning true). The second one is suppressed when the user dismissed
-// the first via DismissSessionWarning. The daemon turns the same events into
-// its tray notification.
-type StateChangeListener interface {
-	OnStateChanged()
-	OnSessionExpiring(expiresAtUnix int64, leadMinutes int64, finalWarning bool)
-}
-
-// Status returns the connect run-loop's status label — the same value the
-// desktop daemon serves in StatusResponse.Status. "NeedsLogin" means the
-// management server rejected the peer and an interactive login is required.
-//
-// The label is latched: the run loop keeps its status in a per-run context
-// state, which a restart replaces with a fresh Idle one, so an engine restart
-// (network change, always-on) would otherwise erase the fact that the peer
-// still needs to log in. Only a successful interactive login or extend clears
-// it — see clearLoginRequired.
-func (c *Client) Status() string {
-	latched, generation := c.loginRequiredState()
-	if latched {
-		return string(internal.StatusNeedsLogin)
-	}
-	cc := c.getConnectClient()
-	if cc == nil {
-		return string(internal.StatusIdle)
-	}
-	status := cc.Status()
-	if status == internal.StatusNeedsLogin {
-		c.latchLoginRequired(generation)
-	}
-	return string(status)
-}
-
-func (c *Client) loginRequiredState() (bool, uint64) {
-	c.loginRequiredMu.Lock()
-	defer c.loginRequiredMu.Unlock()
-	return c.loginRequired, c.loginCleared
-}
-
-// latchLoginRequired records a NeedsLogin observation, unless a clear landed
-// while the caller was reading the run loop's status: cc.Status() is read
-// outside the lock, so a login or extend completing in that window would
-// otherwise be undone by this stale observation, stranding the UI on
-// "login required" over a healthy session.
-func (c *Client) latchLoginRequired(observedGeneration uint64) {
-	c.loginRequiredMu.Lock()
-	defer c.loginRequiredMu.Unlock()
-	if c.loginCleared != observedGeneration {
-		return
-	}
-	c.loginRequired = true
-}
-
-// clearLoginRequired releases the latch after a successful interactive login
-// or session extend, and invalidates any observation already in flight.
-func (c *Client) clearLoginRequired() {
-	c.loginRequiredMu.Lock()
-	defer c.loginRequiredMu.Unlock()
-	c.loginRequired = false
-	c.loginCleared++
-}
-
-// SessionExpiresAtUnix returns the SSO session deadline as unix seconds, or 0
-// when no deadline is known (not SSO-registered, expiry disabled, or the
-// engine has not received one yet). A past value means the session expired.
-// Mirror of StatusResponse.sessionExpiresAt on the desktop daemon.
-func (c *Client) SessionExpiresAtUnix() int64 {
-	deadline := c.recorder.GetSessionExpiresAt()
-	if deadline.IsZero() {
-		return 0
-	}
-	return deadline.Unix()
-}
-
-// SetStateChangeListener registers the state notification listener.
-// Replaces any previously registered listener; remove it with
-// RemoveStateChangeListener.
-func (c *Client) SetStateChangeListener(listener StateChangeListener) {
-	c.stateChangeMu.Lock()
-	defer c.stateChangeMu.Unlock()
-	c.stopStateChangeWatchLocked()
-	if listener == nil {
-		return
-	}
-
-	// Both subscriptions are buffered (one pending tick, ten pending events),
-	// so unsubscribing is not enough to stop callbacks: the loops would drain
-	// what is already queued and deliver it to a listener the caller has
-	// already removed or replaced. Gate every callback on this registration's
-	// own signal, which is closed before unsubscribing.
-	done := make(chan struct{})
-	c.stateChangeDone = done
-
-	id, ch := c.recorder.SubscribeToStateChanges()
-	c.stateChangeSubID = id
-	// The channel is closed by UnsubscribeFromStateChanges, which ends the
-	// goroutine. Ticks are coalesced (buffer of one), so a burst of changes
-	// wakes the listener once.
-	go func() {
-		for range ch {
-			select {
-			case <-done:
-				return
-			default:
-			}
-			listener.OnStateChanged()
-		}
-	}()
-
-	c.eventSub = c.recorder.SubscribeToEvents()
-	go watchSessionWarnings(c.eventSub, listener, done)
-}
-
-// RemoveStateChangeListener unregisters the state notification listener.
-func (c *Client) RemoveStateChangeListener() {
-	c.stateChangeMu.Lock()
-	defer c.stateChangeMu.Unlock()
-	c.stopStateChangeWatchLocked()
-}
-
-// DismissSessionWarning records the user's "Dismiss" on the first expiry
-// warning and suppresses the final one for the current deadline. A refreshed
-// deadline re-arms both. No-op while the engine is not running.
-func (c *Client) DismissSessionWarning() {
-	cc := c.getConnectClient()
-	if cc == nil {
-		return
-	}
-	engine := cc.Engine()
-	if engine == nil {
-		return
-	}
-	engine.DismissSessionWarning()
-}
-
-// ExtendAuthSession runs the interactive SSO flow to obtain a fresh JWT and
-// asks the management server to extend the session deadline. The tunnel is
-// untouched: no resync, no reconnect. Async; the result arrives on the
-// listener. Mirror of the daemon's RequestExtendAuthSession /
-// WaitExtendAuthSession RPC pair, with URLOpener playing the "UI opens the
-// browser" role.
-//
-// Only one flow may be in flight: the PKCE step binds a fixed loopback port,
-// so a second concurrent flow would fail on that bind. Call
-// CancelExtendAuthSession when the user abandons the browser.
-func (c *Client) ExtendAuthSession(urlOpener URLOpener, isAndroidTV bool, resultListener ErrListener) {
-	ctx, err := c.beginExtend()
-	if err != nil {
-		resultListener.OnError(err)
-		return
-	}
-
-	go func() {
-		defer c.endExtend()
-		if err := c.extendAuthSession(ctx, urlOpener, isAndroidTV); err != nil {
-			resultListener.OnError(err)
-			return
-		}
-		resultListener.OnSuccess()
-	}()
-}
-
-// CancelExtendAuthSession aborts an in-flight ExtendAuthSession. The tunnel is
-// left alone — unlike the login flow, which cancels the whole client context
-// by stopping the engine. Without this the abandoned PKCE wait keeps its
-// loopback port for the full flow timeout and blocks every later attempt.
-// No-op when no flow is running.
-func (c *Client) CancelExtendAuthSession() {
-	c.extendMu.Lock()
-	defer c.extendMu.Unlock()
-	if c.extendCancel != nil {
-		c.extendCancel()
-	}
-}
-
-func (c *Client) stopStateChangeWatchLocked() {
-	// Signal first, unsubscribe second: closing the channels only stops new
-	// items, and the loops would still hand whatever is buffered to a listener
-	// that is no longer registered.
-	if c.stateChangeDone != nil {
-		close(c.stateChangeDone)
-		c.stateChangeDone = nil
-	}
-	if c.stateChangeSubID != "" {
-		c.recorder.UnsubscribeFromStateChanges(c.stateChangeSubID)
-		c.stateChangeSubID = ""
-	}
-	if c.eventSub != nil {
-		// Closes the channel, which ends watchSessionWarnings.
-		c.recorder.UnsubscribeFromEvents(c.eventSub)
-		c.eventSub = nil
-	}
-}
-
-// watchSessionWarnings forwards the engine's session-expiry warnings to the
-// listener. The event stream also carries unrelated traffic — network-map
-// updates on every sync, DNS and route errors — so everything but an
-// AUTHENTICATION event carrying the session-warning marker is dropped. Exits
-// when the subscription is closed by UnsubscribeFromEvents, or earlier when
-// done is closed — the stream buffers up to ten events, and a deregistered
-// listener must not receive the ones already queued.
-func watchSessionWarnings(sub *peer.EventSubscription, listener StateChangeListener, done <-chan struct{}) {
-	for ev := range sub.Events() {
-		select {
-		case <-done:
-			return
-		default:
-		}
-		if ev.GetCategory() != cProto.SystemEvent_AUTHENTICATION {
-			continue
-		}
-		meta := ev.GetMetadata()
-		if meta[sessionwatch.MetaSessionWarning] != "true" {
-			// Other AUTHENTICATION events exist (e.g. a deadline rejected as
-			// out of range); they carry no warning marker.
-			continue
-		}
-		deadline, err := sessionwatch.ParseExpiresAt(meta[sessionwatch.MetaSessionExpiresAt])
-		if err != nil {
-			log.Warnf("session warning event with unparsable deadline: %v", err)
-			continue
-		}
-		lead, err := sessionwatch.ParseLeadMinutes(meta[sessionwatch.MetaSessionLeadMinutes])
-		if err != nil {
-			// Informational only — the deadline above is what drives the UI.
-			lead = 0
-		}
-		listener.OnSessionExpiring(deadline.Unix(), int64(lead),
-			meta[sessionwatch.MetaSessionFinal] == "true")
-	}
-}
-
-func (c *Client) beginExtend() (context.Context, error) {
-	c.extendMu.Lock()
-	defer c.extendMu.Unlock()
-	if c.extendCancel != nil {
-		return nil, fmt.Errorf("session extend already in progress")
-	}
-	ctx, cancel := context.WithCancel(context.Background())
-	c.extendCancel = cancel
-	return ctx, nil
-}
-
-func (c *Client) endExtend() {
-	c.extendMu.Lock()
-	defer c.extendMu.Unlock()
-	if c.extendCancel != nil {
-		c.extendCancel()
-		c.extendCancel = nil
-	}
-}
-
-func (c *Client) extendAuthSession(ctx context.Context, urlOpener URLOpener, isAndroidTV bool) error {
-	cfg, _, cc := c.stateSnapshot()
-	if cfg == nil || cc == nil {
-		return fmt.Errorf("engine is not running")
-	}
-	engine := cc.Engine()
-	if engine == nil {
-		return fmt.Errorf("engine is not initialized")
-	}
-
-	authClient, err := auth.NewAuth(ctx, cfg.PrivateKey, cfg.ManagementURL, cfg)
-	if err != nil {
-		return fmt.Errorf("failed to create auth client: %v", err)
-	}
-	defer authClient.Close()
-
-	a := &Auth{ctx: ctx, config: cfg}
-	tokenInfo, err := a.foregroundGetTokenInfo(authClient, urlOpener, isAndroidTV)
-	if err != nil {
-		return fmt.Errorf("interactive sso login failed: %v", err)
-	}
-
-	if _, err := engine.ExtendAuthSession(ctx, tokenInfo.GetTokenToUse()); err != nil {
-		return err
-	}
-	c.clearLoginRequired()
-
-	go urlOpener.OnLoginSuccess()
-	return nil
-}

client/cmd/login.go

@@ -17,26 +17,16 @@ import (
 	"github.com/netbirdio/netbird/client/internal"
 	"github.com/netbirdio/netbird/client/internal/auth"
 	"github.com/netbirdio/netbird/client/internal/profilemanager"
-	nbnet "github.com/netbirdio/netbird/client/net"
 	"github.com/netbirdio/netbird/client/proto"
-	"github.com/netbirdio/netbird/client/server"
 	"github.com/netbirdio/netbird/client/system"
 	"github.com/netbirdio/netbird/util"
 )
 
-// extendSessionFlag drives the `netbird login --extend` flow: refresh the
-// SSO session expiry on the management server without tearing down the
-// tunnel. Mutually exclusive with setup-key login (a setup-key cannot
-// refresh an SSO-tracked peer — see auth.errSetupKeyOnSSOExpiredPeer).
-var extendSessionFlag bool
-
 func init() {
 	loginCmd.PersistentFlags().BoolVar(&noBrowser, noBrowserFlag, false, noBrowserDesc)
 	loginCmd.PersistentFlags().BoolVar(&showQR, showQRFlag, false, showQRDesc)
 	loginCmd.PersistentFlags().StringVar(&profileName, profileNameFlag, "", profileNameDesc)
 	loginCmd.PersistentFlags().StringVarP(&configPath, "config", "c", "", "(DEPRECATED) Netbird config file location")
-	loginCmd.PersistentFlags().BoolVar(&extendSessionFlag, "extend", false,
-		"refresh the SSO session expiry without tearing down the tunnel (requires an active connection)")
 }
 
 var loginCmd = &cobra.Command{
@@ -71,16 +61,6 @@ var loginCmd = &cobra.Command{
 			return err
 		}
 
-		if extendSessionFlag {
-			if providedSetupKey != "" {
-				return fmt.Errorf("--extend cannot be combined with a setup key; setup keys can only enrol new peers")
-			}
-			if err := doExtendSession(ctx, cmd); err != nil {
-				return fmt.Errorf("extend session failed: %v", err)
-			}
-			return nil
-		}
-
 		// workaround to run without service
 		if util.FindFirstLogPath(logFiles) == "" {
 			if err := doForegroundLogin(ctx, cmd, providedSetupKey, activeProf); err != nil {
@@ -172,65 +152,6 @@ func doDaemonLogin(ctx context.Context, cmd *cobra.Command, providedSetupKey str
 	return nil
 }
 
-// doExtendSession drives the daemon's RequestExtendAuthSession /
-// WaitExtendAuthSession pair. The user is sent through a regular SSO flow
-// (browser + verification URL) and the resulting JWT is forwarded to the
-// management server's ExtendAuthSession RPC. The tunnel stays up
-// throughout — no Down/Up, no network-map resync.
-func doExtendSession(ctx context.Context, cmd *cobra.Command) 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()
-
-	client := proto.NewDaemonServiceClient(conn)
-
-	req := &proto.RequestExtendAuthSessionRequest{}
-	// Pre-fill the IdP login hint from the active profile so the user
-	// doesn't have to retype their email. Best-effort: we still proceed
-	// without a hint if the lookup fails.
-	pm := profilemanager.NewProfileManager()
-	if active, perr := pm.GetActiveProfile(); perr == nil {
-		if profState, sperr := pm.GetProfileState(active.ID); sperr == nil && profState.Email != "" {
-			req.Hint = &profState.Email
-		}
-	}
-
-	startResp, err := client.RequestExtendAuthSession(ctx, req)
-	if err != nil {
-		return fmt.Errorf("start extend session: %v", err)
-	}
-
-	uri := startResp.GetVerificationURIComplete()
-	if uri == "" {
-		uri = startResp.GetVerificationURI()
-	}
-	openURL(cmd, uri, startResp.GetUserCode(), noBrowser, showQR)
-
-	waitResp, err := client.WaitExtendAuthSession(ctx, &proto.WaitExtendAuthSessionRequest{
-		DeviceCode: startResp.GetDeviceCode(),
-		UserCode:   startResp.GetUserCode(),
-	})
-	if err != nil {
-		return fmt.Errorf("wait for extend session: %v", err)
-	}
-
-	if ts := waitResp.GetSessionExpiresAt(); ts.IsValid() && !ts.AsTime().IsZero() {
-		deadline := ts.AsTime().Local()
-		cmd.Printf("Session extended. New expiry: %s\n", deadline.Format("2006-01-02 15:04:05 MST"))
-	} else {
-		// Management reported the peer is not eligible (e.g. login
-		// expiration disabled on the account). Surface that fact
-		// instead of pretending the call succeeded.
-		cmd.Println("Session extension call completed, but the management server did not return a new deadline (peer may not be SSO-tracked or login expiration is disabled).")
-	}
-	return nil
-}
-
 func getActiveProfile(ctx context.Context, pm *profilemanager.ProfileManager, profileName string, username string) (*profilemanager.Profile, error) {
 	// switch profile if provided
 
@@ -333,14 +254,6 @@ func doForegroundLogin(ctx context.Context, cmd *cobra.Command, setupKey string,
 		return fmt.Errorf("read config file %s: %v", configFilePath, err)
 	}
 
-	// Mirror runInForegroundMode: recover residual state (DNS, firewall,
-	// ssh config, legacy routing) from a previous unclean shutdown and
-	// enable advanced routing before dialing management.
-	if err := server.RestoreResidualState(ctx, profilemanager.NewServiceManager(configFilePath).GetStatePath()); err != nil {
-		log.Warnf("failed to restore residual state: %v", err)
-	}
-	nbnet.Init()
-
 	err = foregroundLogin(ctx, cmd, config, setupKey, activeProf.ID)
 	if err != nil {
 		return fmt.Errorf("foreground login failed: %v", err)

client/cmd/service.go

@@ -5,7 +5,6 @@ package cmd
 import (
 	"context"
 	"fmt"
-	"net/http"
 	"runtime"
 	"strings"
 	"sync"
@@ -23,21 +22,15 @@ var serviceCmd = &cobra.Command{
 	Short: "Manage the NetBird daemon service",
 }
 
-const defaultJSONSocket = "unix:///var/run/netbird-http.sock"
-
 var (
-	serviceName      string
+	serviceName    string
-	serviceEnvVars   []string
+	serviceEnvVars []string
-	jsonSocket       string
-	enableJSONSocket bool
 )
 
 type program struct {
 	ctx              context.Context
 	cancel           context.CancelFunc
 	serv             *grpc.Server
-	jsonServ         *http.Server
-	jsonServMu       sync.Mutex
 	serverInstance   *server.Server
 	serverInstanceMu sync.Mutex
 }
@@ -53,8 +46,6 @@ func init() {
 	serviceCmd.PersistentFlags().BoolVar(&updateSettingsDisabled, "disable-update-settings", false, "Disables update settings feature. If enabled, the client will not be able to change or edit any settings. To persist this setting, use: netbird service install --disable-update-settings")
 	serviceCmd.PersistentFlags().BoolVar(&captureEnabled, "enable-capture", false, "Enables packet capture via 'netbird debug capture'. To persist, use: netbird service install --enable-capture")
 	serviceCmd.PersistentFlags().BoolVar(&networksDisabled, "disable-networks", false, "Disables network selection. If enabled, the client will not allow listing, selecting, or deselecting networks. To persist, use: netbird service install --disable-networks")
-	serviceCmd.PersistentFlags().BoolVar(&enableJSONSocket, "enable-json-socket", false, "Enables the HTTP/JSON API socket served by grpc-gateway. To persist, use: netbird service install --enable-json-socket")
-	serviceCmd.PersistentFlags().StringVar(&jsonSocket, "json-socket", defaultJSONSocket, "HTTP/JSON API socket address [unix|tcp]://[path|host:port]. Requires --enable-json-socket to serve. To persist, use: netbird service install --enable-json-socket --json-socket")
 
 	rootCmd.PersistentFlags().StringVarP(&serviceName, "service", "s", defaultServiceName, "Netbird system service name")
 	serviceEnvDesc := `Sets extra environment variables for the service. ` +

client/cmd/service_controller.go

@@ -5,6 +5,9 @@ package cmd
 import (
 	"context"
 	"fmt"
+	"net"
+	"os"
+	"strings"
 	"time"
 
 	"github.com/kardianos/service"
@@ -19,56 +22,41 @@ import (
 	"github.com/netbirdio/netbird/util"
 )
 
-func validateJSONSocketFlags() error {
-	if serviceCmd.PersistentFlags().Changed("json-socket") && !enableJSONSocket {
-		return fmt.Errorf("--json-socket requires --enable-json-socket to configure the daemon JSON gateway")
-	}
-	return nil
-}
-
 func (p *program) Start(svc service.Service) error {
 	// Start should not block. Do the actual work async.
 	log.Info("starting NetBird service") //nolint
 
-	if err := validateJSONSocketFlags(); err != nil {
-		return err
-	}
-
 	// Collect static system and platform information
 	system.UpdateStaticInfoAsync()
 
 	// in any case, even if configuration does not exists we run daemon to serve CLI gRPC API.
 	p.serv = grpc.NewServer()
 
-	daemonListener, err := listenOnAddress(daemonAddr)
+	split := strings.Split(daemonAddr, "://")
+	switch split[0] {
+	case "unix":
+		// cleanup failed close
+		stat, err := os.Stat(split[1])
+		if err == nil && !stat.IsDir() {
+			if err := os.Remove(split[1]); err != nil {
+				log.Debugf("remove socket file: %v", err)
+			}
+		}
+	case "tcp":
+	default:
+		return fmt.Errorf("unsupported daemon address protocol: %v", split[0])
+	}
+
+	listen, err := net.Listen(split[0], split[1])
 	if err != nil {
 		return fmt.Errorf("listen daemon interface: %w", err)
 	}
-
-	var jsonListener *socketListener
-	if enableJSONSocket {
-		jsonListener, err = listenOnAddress(jsonSocket)
-		if err != nil {
-			_ = daemonListener.Close()
-			return fmt.Errorf("listen daemon JSON interface: %w", err)
-		}
-	} else {
-		removeStaleUnixSocketForAddress(jsonSocket)
-	}
-
 	go func() {
-		defer daemonListener.Close()
+		defer listen.Close()
-		if jsonListener != nil {
-			defer jsonListener.Close()
-		}
 
-		if err := daemonListener.chmodUnixSocket("daemon"); err != nil {
+		if split[0] == "unix" {
-			log.Error(err)
+			if err := os.Chmod(split[1], 0666); err != nil {
-			return
+				log.Errorf("failed setting daemon permissions: %v", split[1])
-		}
-		if jsonListener != nil {
-			if err := jsonListener.chmodUnixSocket("daemon JSON"); err != nil {
-				log.Error(err)
 				return
 			}
 		}
@@ -83,16 +71,8 @@ func (p *program) Start(svc service.Service) error {
 		p.serverInstance = serverInstance
 		p.serverInstanceMu.Unlock()
 
-		if jsonListener != nil {
+		log.Printf("started daemon server: %v", split[1])
-			if err := p.startJSONGateway(jsonListener, daemonAddr); err != nil {
+		if err := p.serv.Serve(listen); err != nil {
-				log.Fatalf("failed to start daemon JSON server: %v", err)
-			}
-		} else {
-			log.Debug("daemon JSON socket disabled")
-		}
-
-		log.Printf("started daemon server: %v", daemonListener.address)
-		if err := p.serv.Serve(daemonListener.Listener); err != nil {
 			log.Errorf("failed to serve daemon requests: %v", err)
 		}
 	}()
@@ -112,20 +92,6 @@ func (p *program) Stop(srv service.Service) error {
 
 	p.cancel()
 
-	p.jsonServMu.Lock()
-	jsonServ := p.jsonServ
-	p.jsonServMu.Unlock()
-	if jsonServ != nil {
-		shutdownCtx, shutdownCancel := context.WithTimeout(context.Background(), 2*time.Second)
-		if err := jsonServ.Shutdown(shutdownCtx); err != nil {
-			log.Errorf("failed to stop daemon JSON server gracefully: %v", err)
-			if err := jsonServ.Close(); err != nil {
-				log.Errorf("failed to close daemon JSON server: %v", err)
-			}
-		}
-		shutdownCancel()
-	}
-
 	if p.serv != nil {
 		p.serv.Stop()
 	}
@@ -182,9 +148,6 @@ var runCmd = &cobra.Command{
 		if err != nil {
 			return err
 		}
-		if err := validateJSONSocketFlags(); err != nil {
-			return err
-		}
 
 		return s.Run()
 	},
@@ -199,9 +162,6 @@ var startCmd = &cobra.Command{
 		if err != nil {
 			return err
 		}
-		if err := validateJSONSocketFlags(); err != nil {
-			return err
-		}
 
 		if err := s.Start(); err != nil {
 			return fmt.Errorf("start service: %w", err)
@@ -238,9 +198,6 @@ var restartCmd = &cobra.Command{
 		if err != nil {
 			return err
 		}
-		if err := validateJSONSocketFlags(); err != nil {
-			return err
-		}
 
 		if err := s.Restart(); err != nil {
 			return fmt.Errorf("restart service: %w", err)

client/cmd/service_installer.go

@@ -67,10 +67,6 @@ func buildServiceArguments() []string {
 		args = append(args, "--disable-networks")
 	}
 
-	if enableJSONSocket {
-		args = append(args, "--enable-json-socket", "--json-socket", jsonSocket)
-	}
-
 	return args
 }
 
@@ -110,10 +106,6 @@ func configurePlatformSpecificSettings(svcConfig *service.Config) error {
 
 // Create fully configured service config for install/reconfigure
 func createServiceConfigForInstall() (*service.Config, error) {
-	if err := validateJSONSocketFlags(); err != nil {
-		return nil, err
-	}
-
 	svcConfig, err := newSVCConfig()
 	if err != nil {
 		return nil, fmt.Errorf("create service config: %w", err)

client/cmd/service_json_gateway.go

@@ -1,52 +0,0 @@
-//go:build !ios && !android
-
-package cmd
-
-import (
-	"context"
-	"errors"
-	"net"
-	"net/http"
-	"strings"
-	"time"
-
-	"github.com/grpc-ecosystem/grpc-gateway/v2/runtime"
-	log "github.com/sirupsen/logrus"
-	"google.golang.org/grpc"
-	"google.golang.org/grpc/credentials/insecure"
-
-	"github.com/netbirdio/netbird/client/proto"
-)
-
-func grpcGatewayEndpoint(addr string) string {
-	return strings.TrimPrefix(addr, "tcp://")
-}
-
-func (p *program) startJSONGateway(jsonListener *socketListener, daemonEndpoint string) error {
-	mux := runtime.NewServeMux()
-	opts := []grpc.DialOption{grpc.WithTransportCredentials(insecure.NewCredentials())}
-	if err := proto.RegisterDaemonServiceHandlerFromEndpoint(p.ctx, mux, grpcGatewayEndpoint(daemonEndpoint), opts); err != nil {
-		return err
-	}
-
-	jsonServer := &http.Server{
-		Handler:           mux,
-		ReadHeaderTimeout: 5 * time.Second,
-		BaseContext: func(net.Listener) context.Context {
-			return p.ctx
-		},
-	}
-
-	p.jsonServMu.Lock()
-	p.jsonServ = jsonServer
-	p.jsonServMu.Unlock()
-
-	go func() {
-		log.Printf("started daemon JSON server: %v", jsonListener.address)
-		if err := jsonServer.Serve(jsonListener.Listener); err != nil && !errors.Is(err, http.ErrServerClosed) {
-			log.Errorf("failed to serve daemon JSON requests: %v", err)
-		}
-	}()
-
-	return nil
-}

client/cmd/service_json_socket_test.go

@@ -1,176 +0,0 @@
-//go:build !ios && !android
-
-package cmd
-
-import (
-	"net"
-	"os"
-	"path/filepath"
-	"runtime"
-	"testing"
-
-	"github.com/spf13/cobra"
-	"github.com/spf13/pflag"
-	"github.com/stretchr/testify/assert"
-	"github.com/stretchr/testify/require"
-)
-
-func preserveJSONSocketTestState(t *testing.T) {
-	t.Helper()
-
-	origJSONSocket := jsonSocket
-	origEnableJSONSocket := enableJSONSocket
-	origChanged := map[string]bool{}
-	serviceCmd.PersistentFlags().VisitAll(func(flag *pflag.Flag) {
-		origChanged[flag.Name] = flag.Changed
-	})
-
-	t.Cleanup(func() {
-		jsonSocket = origJSONSocket
-		enableJSONSocket = origEnableJSONSocket
-		serviceCmd.PersistentFlags().VisitAll(func(flag *pflag.Flag) {
-			flag.Changed = origChanged[flag.Name]
-		})
-	})
-}
-
-func TestJSONSocketFlagsArePositiveEnableOnly(t *testing.T) {
-	assert.NotNil(t, serviceCmd.PersistentFlags().Lookup("enable-json-socket"))
-	assert.NotNil(t, serviceCmd.PersistentFlags().Lookup("json-socket"))
-	assert.Nil(t, serviceCmd.PersistentFlags().Lookup("disable-json-socket"))
-	assert.Equal(t, "false", serviceCmd.PersistentFlags().Lookup("enable-json-socket").DefValue)
-}
-
-func TestBuildServiceArgumentsDefaultDisablesJSONSocket(t *testing.T) {
-	preserveJSONSocketTestState(t)
-
-	enableJSONSocket = false
-	jsonSocket = "tcp://127.0.0.1:8080"
-
-	args := buildServiceArguments()
-
-	assert.NotContains(t, args, "--enable-json-socket")
-	assert.NotContains(t, args, "--json-socket")
-}
-
-func TestBuildServiceArgumentsIncludesJSONSocketWhenEnabled(t *testing.T) {
-	preserveJSONSocketTestState(t)
-
-	enableJSONSocket = true
-	jsonSocket = "tcp://127.0.0.1:8080"
-
-	args := buildServiceArguments()
-
-	enableIndex := indexOfArg(args, "--enable-json-socket")
-	jsonIndex := indexOfArg(args, "--json-socket")
-	require.NotEqual(t, -1, enableIndex)
-	require.NotEqual(t, -1, jsonIndex)
-	require.Less(t, enableIndex, jsonIndex)
-	require.Less(t, jsonIndex+1, len(args))
-	assert.Equal(t, "tcp://127.0.0.1:8080", args[jsonIndex+1])
-}
-
-func TestJSONSocketWithoutEnableValidation(t *testing.T) {
-	preserveJSONSocketTestState(t)
-
-	enableJSONSocket = false
-	require.NoError(t, serviceCmd.PersistentFlags().Set("json-socket", "tcp://127.0.0.1:8080"))
-
-	err := validateJSONSocketFlags()
-
-	require.Error(t, err)
-	assert.Contains(t, err.Error(), "--enable-json-socket")
-}
-
-func TestJSONSocketWithEnableValidation(t *testing.T) {
-	preserveJSONSocketTestState(t)
-
-	require.NoError(t, serviceCmd.PersistentFlags().Set("enable-json-socket", "true"))
-	require.NoError(t, serviceCmd.PersistentFlags().Set("json-socket", "tcp://127.0.0.1:8080"))
-
-	assert.NoError(t, validateJSONSocketFlags())
-}
-
-func TestJSONSocketServiceParamsPersistEnableAndAddress(t *testing.T) {
-	preserveJSONSocketTestState(t)
-	serviceCmd.PersistentFlags().VisitAll(func(flag *pflag.Flag) {
-		flag.Changed = false
-	})
-
-	enableJSONSocket = true
-	jsonSocket = "tcp://127.0.0.1:8080"
-
-	params := currentServiceParams()
-	require.True(t, params.EnableJSONSocket)
-	require.Equal(t, "tcp://127.0.0.1:8080", params.JSONSocket)
-
-	enableJSONSocket = false
-	jsonSocket = defaultJSONSocket
-	applyServiceParams(testServiceEnvCommand(), params)
-
-	assert.True(t, enableJSONSocket)
-	assert.Equal(t, "tcp://127.0.0.1:8080", jsonSocket)
-}
-
-func TestRemoveStaleUnixSocketDoesNotRemoveRegularFile(t *testing.T) {
-	path := filepath.Join(t.TempDir(), "netbird-http.sock")
-	require.NoError(t, os.WriteFile(path, []byte("not a socket"), 0600))
-
-	removeStaleUnixSocket(path)
-
-	data, err := os.ReadFile(path)
-	require.NoError(t, err)
-	assert.Equal(t, []byte("not a socket"), data)
-}
-
-func TestRemoveStaleUnixSocketRemovesSocket(t *testing.T) {
-	if runtime.GOOS == "windows" {
-		t.Skip("unix sockets are not available on Windows")
-	}
-
-	path := filepath.Join(t.TempDir(), "netbird-http.sock")
-	addr := &net.UnixAddr{Name: path, Net: "unix"}
-	listener, err := net.ListenUnix("unix", addr)
-	require.NoError(t, err)
-	listener.SetUnlinkOnClose(false)
-	require.NoError(t, listener.Close())
-
-	_, err = os.Lstat(path)
-	require.NoError(t, err, "test setup must leave a stale Unix socket path")
-
-	removeStaleUnixSocket(path)
-
-	_, err = os.Lstat(path)
-	assert.True(t, os.IsNotExist(err), "expected stale Unix socket to be removed, got %v", err)
-}
-
-func TestRemoveStaleUnixSocketDoesNotRemoveLiveSocket(t *testing.T) {
-	if runtime.GOOS == "windows" {
-		t.Skip("unix sockets are not available on Windows")
-	}
-
-	path := filepath.Join(t.TempDir(), "netbird-http.sock")
-	listener, err := net.Listen("unix", path)
-	require.NoError(t, err)
-	defer listener.Close()
-
-	removeStaleUnixSocket(path)
-
-	_, err = os.Lstat(path)
-	assert.NoError(t, err, "expected live Unix socket to be preserved")
-}
-
-func testServiceEnvCommand() *cobra.Command {
-	cmd := &cobra.Command{}
-	cmd.Flags().StringSlice("service-env", nil, "")
-	return cmd
-}
-
-func indexOfArg(args []string, arg string) int {
-	for i, candidate := range args {
-		if candidate == arg {
-			return i
-		}
-	}
-	return -1
-}

client/cmd/service_params.go

@@ -23,7 +23,6 @@ const serviceParamsFile = "service.json"
 type serviceParams struct {
 	LogLevel              string            `json:"log_level"`
 	DaemonAddr            string            `json:"daemon_addr"`
-	JSONSocket            string            `json:"json_socket"`
 	ManagementURL         string            `json:"management_url,omitempty"`
 	ConfigPath            string            `json:"config_path,omitempty"`
 	LogFiles              []string          `json:"log_files,omitempty"`
@@ -31,7 +30,6 @@ type serviceParams struct {
 	DisableUpdateSettings bool              `json:"disable_update_settings,omitempty"`
 	EnableCapture         bool              `json:"enable_capture,omitempty"`
 	DisableNetworks       bool              `json:"disable_networks,omitempty"`
-	EnableJSONSocket      bool              `json:"enable_json_socket,omitempty"`
 	ServiceEnvVars        map[string]string `json:"service_env_vars,omitempty"`
 }
 
@@ -77,7 +75,6 @@ func currentServiceParams() *serviceParams {
 	params := &serviceParams{
 		LogLevel:              logLevel,
 		DaemonAddr:            daemonAddr,
-		JSONSocket:            jsonSocket,
 		ManagementURL:         managementURL,
 		ConfigPath:            configPath,
 		LogFiles:              logFiles,
@@ -85,7 +82,6 @@ func currentServiceParams() *serviceParams {
 		DisableUpdateSettings: updateSettingsDisabled,
 		EnableCapture:         captureEnabled,
 		DisableNetworks:       networksDisabled,
-		EnableJSONSocket:      enableJSONSocket,
 	}
 
 	if len(serviceEnvVars) > 0 {
@@ -117,8 +113,9 @@ func applyServiceParams(cmd *cobra.Command, params *serviceParams) {
 		return
 	}
 
-	// For fields with non-empty defaults, keep the != "" guard so that an older
+	// For fields with non-empty defaults (log-level, daemon-addr), keep the
-	// service.json missing the field doesn't clobber the default with an empty string.
+	// != "" guard so that an older service.json missing the field doesn't
+	// clobber the default with an empty string.
 	if !rootCmd.PersistentFlags().Changed("log-level") && params.LogLevel != "" {
 		logLevel = params.LogLevel
 	}
@@ -127,14 +124,6 @@ func applyServiceParams(cmd *cobra.Command, params *serviceParams) {
 		daemonAddr = params.DaemonAddr
 	}
 
-	if !serviceCmd.PersistentFlags().Changed("json-socket") && params.JSONSocket != "" {
-		jsonSocket = params.JSONSocket
-	}
-
-	if !serviceCmd.PersistentFlags().Changed("enable-json-socket") {
-		enableJSONSocket = params.EnableJSONSocket
-	}
-
 	// For optional fields where empty means "use default", always apply so
 	// that an explicit clear (--management-url "") persists across reinstalls.
 	if !rootCmd.PersistentFlags().Changed("management-url") {

client/cmd/service_params_test.go

@@ -41,8 +41,6 @@ func TestSaveAndLoadServiceParams(t *testing.T) {
 	params := &serviceParams{
 		LogLevel:              "debug",
 		DaemonAddr:            "unix:///var/run/netbird.sock",
-		JSONSocket:            "tcp://127.0.0.1:8080",
-		EnableJSONSocket:      true,
 		ManagementURL:         "https://my.server.com",
 		ConfigPath:            "/etc/netbird/config.json",
 		LogFiles:              []string{"/var/log/netbird/client.log", "console"},
@@ -65,8 +63,6 @@ func TestSaveAndLoadServiceParams(t *testing.T) {
 
 	assert.Equal(t, params.LogLevel, loaded.LogLevel)
 	assert.Equal(t, params.DaemonAddr, loaded.DaemonAddr)
-	assert.Equal(t, params.JSONSocket, loaded.JSONSocket)
-	assert.Equal(t, params.EnableJSONSocket, loaded.EnableJSONSocket)
 	assert.Equal(t, params.ManagementURL, loaded.ManagementURL)
 	assert.Equal(t, params.ConfigPath, loaded.ConfigPath)
 	assert.Equal(t, params.LogFiles, loaded.LogFiles)
@@ -105,8 +101,6 @@ func TestLoadServiceParams_InvalidJSON(t *testing.T) {
 func TestCurrentServiceParams(t *testing.T) {
 	origLogLevel := logLevel
 	origDaemonAddr := daemonAddr
-	origJSONSocket := jsonSocket
-	origEnableJSONSocket := enableJSONSocket
 	origManagementURL := managementURL
 	origConfigPath := configPath
 	origLogFiles := logFiles
@@ -116,8 +110,6 @@ func TestCurrentServiceParams(t *testing.T) {
 	t.Cleanup(func() {
 		logLevel = origLogLevel
 		daemonAddr = origDaemonAddr
-		jsonSocket = origJSONSocket
-		enableJSONSocket = origEnableJSONSocket
 		managementURL = origManagementURL
 		configPath = origConfigPath
 		logFiles = origLogFiles
@@ -128,8 +120,6 @@ func TestCurrentServiceParams(t *testing.T) {
 
 	logLevel = "trace"
 	daemonAddr = "tcp://127.0.0.1:9999"
-	jsonSocket = "tcp://127.0.0.1:8080"
-	enableJSONSocket = true
 	managementURL = "https://mgmt.example.com"
 	configPath = "/tmp/test-config.json"
 	logFiles = []string{"/tmp/test.log"}
@@ -141,8 +131,6 @@ func TestCurrentServiceParams(t *testing.T) {
 
 	assert.Equal(t, "trace", params.LogLevel)
 	assert.Equal(t, "tcp://127.0.0.1:9999", params.DaemonAddr)
-	assert.Equal(t, "tcp://127.0.0.1:8080", params.JSONSocket)
-	assert.True(t, params.EnableJSONSocket)
 	assert.Equal(t, "https://mgmt.example.com", params.ManagementURL)
 	assert.Equal(t, "/tmp/test-config.json", params.ConfigPath)
 	assert.Equal(t, []string{"/tmp/test.log"}, params.LogFiles)
@@ -154,8 +142,6 @@ func TestCurrentServiceParams(t *testing.T) {
 func TestApplyServiceParams_OnlyUnchangedFlags(t *testing.T) {
 	origLogLevel := logLevel
 	origDaemonAddr := daemonAddr
-	origJSONSocket := jsonSocket
-	origEnableJSONSocket := enableJSONSocket
 	origManagementURL := managementURL
 	origConfigPath := configPath
 	origLogFiles := logFiles
@@ -165,8 +151,6 @@ func TestApplyServiceParams_OnlyUnchangedFlags(t *testing.T) {
 	t.Cleanup(func() {
 		logLevel = origLogLevel
 		daemonAddr = origDaemonAddr
-		jsonSocket = origJSONSocket
-		enableJSONSocket = origEnableJSONSocket
 		managementURL = origManagementURL
 		configPath = origConfigPath
 		logFiles = origLogFiles
@@ -178,8 +162,6 @@ func TestApplyServiceParams_OnlyUnchangedFlags(t *testing.T) {
 	// Reset all flags to defaults.
 	logLevel = "info"
 	daemonAddr = "unix:///var/run/netbird.sock"
-	jsonSocket = defaultJSONSocket
-	enableJSONSocket = false
 	managementURL = ""
 	configPath = "/etc/netbird/config.json"
 	logFiles = []string{"/var/log/netbird/client.log"}
@@ -202,8 +184,6 @@ func TestApplyServiceParams_OnlyUnchangedFlags(t *testing.T) {
 	saved := &serviceParams{
 		LogLevel:              "debug",
 		DaemonAddr:            "tcp://127.0.0.1:5555",
-		JSONSocket:            "tcp://127.0.0.1:8080",
-		EnableJSONSocket:      true,
 		ManagementURL:         "https://saved.example.com",
 		ConfigPath:            "/saved/config.json",
 		LogFiles:              []string{"/saved/client.log"},
@@ -221,8 +201,6 @@ func TestApplyServiceParams_OnlyUnchangedFlags(t *testing.T) {
 
 	// All other fields were not Changed, so they should use saved values.
 	assert.Equal(t, "tcp://127.0.0.1:5555", daemonAddr)
-	assert.Equal(t, "tcp://127.0.0.1:8080", jsonSocket)
-	assert.True(t, enableJSONSocket)
 	assert.Equal(t, "https://saved.example.com", managementURL)
 	assert.Equal(t, "/saved/config.json", configPath)
 	assert.Equal(t, []string{"/saved/client.log"}, logFiles)
@@ -234,17 +212,14 @@ func TestApplyServiceParams_OnlyUnchangedFlags(t *testing.T) {
 func TestApplyServiceParams_BooleanRevertToFalse(t *testing.T) {
 	origProfilesDisabled := profilesDisabled
 	origUpdateSettingsDisabled := updateSettingsDisabled
-	origEnableJSONSocket := enableJSONSocket
 	t.Cleanup(func() {
 		profilesDisabled = origProfilesDisabled
 		updateSettingsDisabled = origUpdateSettingsDisabled
-		enableJSONSocket = origEnableJSONSocket
 	})
 
 	// Simulate current state where booleans are true (e.g. set by previous install).
 	profilesDisabled = true
 	updateSettingsDisabled = true
-	enableJSONSocket = true
 
 	// Reset Changed state so flags appear unset.
 	serviceCmd.PersistentFlags().VisitAll(func(f *pflag.Flag) {
@@ -263,7 +238,6 @@ func TestApplyServiceParams_BooleanRevertToFalse(t *testing.T) {
 
 	assert.False(t, profilesDisabled, "saved false should override current true")
 	assert.False(t, updateSettingsDisabled, "saved false should override current true")
-	assert.False(t, enableJSONSocket, "saved false should override current true")
 }
 
 func TestApplyServiceParams_ClearManagementURL(t *testing.T) {
@@ -556,7 +530,6 @@ func fieldToGlobalVar(field string) string {
 	m := map[string]string{
 		"LogLevel":              "logLevel",
 		"DaemonAddr":            "daemonAddr",
-		"JSONSocket":            "jsonSocket",
 		"ManagementURL":         "managementURL",
 		"ConfigPath":            "configPath",
 		"LogFiles":              "logFiles",
@@ -564,7 +537,6 @@ func fieldToGlobalVar(field string) string {
 		"DisableUpdateSettings": "updateSettingsDisabled",
 		"EnableCapture":         "captureEnabled",
 		"DisableNetworks":       "networksDisabled",
-		"EnableJSONSocket":      "enableJSONSocket",
 		"ServiceEnvVars":        "serviceEnvVars",
 	}
 	if v, ok := m[field]; ok {

client/cmd/service_socket.go

@@ -1,111 +0,0 @@
-//go:build !ios && !android
-
-package cmd
-
-import (
-	"errors"
-	"fmt"
-	"net"
-	"os"
-	"strings"
-	"syscall"
-	"time"
-
-	log "github.com/sirupsen/logrus"
-)
-
-type socketListener struct {
-	net.Listener
-	network string
-	address string
-}
-
-func listenOnAddress(addr string) (*socketListener, error) {
-	network, address, err := parseListenAddress(addr)
-	if err != nil {
-		return nil, err
-	}
-
-	if network == "unix" {
-		removeStaleUnixSocket(address)
-	}
-
-	listener, err := net.Listen(network, address)
-	if err != nil {
-		return nil, err
-	}
-
-	return &socketListener{Listener: listener, network: network, address: address}, nil
-}
-
-func parseListenAddress(addr string) (string, string, error) {
-	network, address, ok := strings.Cut(addr, "://")
-	if !ok || network == "" || address == "" {
-		return "", "", fmt.Errorf("address must be in [unix|tcp]://[path|host:port] format: %q", addr)
-	}
-
-	switch network {
-	case "unix", "tcp":
-		return network, address, nil
-	default:
-		return "", "", fmt.Errorf("unsupported daemon address protocol: %v", network)
-	}
-}
-
-func removeStaleUnixSocket(path string) {
-	stat, err := os.Lstat(path)
-	if err != nil {
-		if !os.IsNotExist(err) {
-			log.Debugf("stat socket file: %v", err)
-		}
-		return
-	}
-
-	if stat.Mode()&os.ModeSocket == 0 {
-		return
-	}
-
-	if !isStaleUnixSocket(path) {
-		return
-	}
-
-	if err := os.Remove(path); err != nil {
-		log.Debugf("remove socket file: %v", err)
-	}
-}
-
-func isStaleUnixSocket(path string) bool {
-	conn, err := net.DialTimeout("unix", path, 100*time.Millisecond)
-	if err == nil {
-		if closeErr := conn.Close(); closeErr != nil {
-			log.Debugf("close unix socket probe: %v", closeErr)
-		}
-		return false
-	}
-
-	if os.IsNotExist(err) || os.IsPermission(err) || os.IsTimeout(err) {
-		log.Debugf("not removing unix socket %s after probe error: %v", path, err)
-		return false
-	}
-
-	return errors.Is(err, syscall.ECONNREFUSED)
-}
-
-func removeStaleUnixSocketForAddress(addr string) {
-	network, address, err := parseListenAddress(addr)
-	if err != nil || network != "unix" {
-		return
-	}
-	removeStaleUnixSocket(address)
-}
-
-func (l *socketListener) chmodUnixSocket(description string) error {
-	if l == nil || l.network != "unix" {
-		return nil
-	}
-
-	if err := os.Chmod(l.address, 0666); err != nil {
-		return fmt.Errorf("failed setting %s permissions for %s: %w", description, l.address, err)
-	}
-	return nil
-}

client/cmd/status.go

@@ -6,7 +6,6 @@ import (
 	"net"
 	"net/netip"
 	"strings"
-	"time"
 
 	"github.com/spf13/cobra"
 	"google.golang.org/grpc/status"
@@ -116,11 +115,6 @@ func statusFunc(cmd *cobra.Command, args []string) error {
 	// manager only knows the active profile ID, not its display name.
 	profName := getActiveProfileName(ctx)
 
-	var sessionExpiresAt time.Time
-	if ts := resp.GetSessionExpiresAt(); ts.IsValid() {
-		sessionExpiresAt = ts.AsTime().UTC()
-	}
-
 	var outputInformationHolder = nbstatus.ConvertToStatusOutputOverview(resp.GetFullStatus(), nbstatus.ConvertOptions{
 		Anonymize:            anonymizeFlag,
 		DaemonVersion:        resp.GetDaemonVersion(),
@@ -131,7 +125,6 @@ func statusFunc(cmd *cobra.Command, args []string) error {
 		IPsFilter:            ipsFilterMap,
 		ConnectionTypeFilter: connectionTypeFilter,
 		ProfileName:          profName,
-		SessionExpiresAt:     sessionExpiresAt,
 	})
 	var statusOutputString string
 	switch {

client/cmd/up.go

@@ -22,8 +22,6 @@ import (
 	"github.com/netbirdio/netbird/client/internal/peer"
 	"github.com/netbirdio/netbird/client/internal/profilemanager"
 	"github.com/netbirdio/netbird/client/proto"
-	nbnet "github.com/netbirdio/netbird/client/net"
-	"github.com/netbirdio/netbird/client/server"
 	"github.com/netbirdio/netbird/client/system"
 	"github.com/netbirdio/netbird/shared/management/domain"
 	"github.com/netbirdio/netbird/util"
@@ -231,24 +229,6 @@ func runInForegroundMode(ctx context.Context, cmd *cobra.Command, activeProf *pr
 
 	_, _ = profilemanager.UpdateOldManagementURL(ctx, config, configFilePath)
 
-	// Restore residual state left by a previous run that did not shut down
-	// cleanly, mirroring what the daemon does before connecting: it recovers
-	// DNS config (a stale resolv.conf takeover can make the management
-	// hostname unresolvable), firewall rules, ssh config and legacy routing.
-	// Route cleanup itself happens at engine start; nbnet.Init() below lets
-	// the management dial bypass a leftover fwmark rule until then.
-	// Foreground mode is particularly exposed in containers: a crashed
-	// container restarts inside the same (pod) network namespace, so stale
-	// state survives while the process does not.
-	if err := server.RestoreResidualState(ctx, profilemanager.NewServiceManager(configPath).GetStatePath()); err != nil {
-		log.Warnf("failed to restore residual state: %v", err)
-	}
-
-	// Enable advanced routing (as the daemon does on startup) so the
-	// management dial bypasses a leftover fwmark rule instead of being
-	// shunted into a stale routing table.
-	nbnet.Init()
-
 	err = foregroundLogin(ctx, cmd, config, providedSetupKey, activeProf.ID)
 	if err != nil {
 		return fmt.Errorf("foreground login failed: %v", err)

client/embed/embed.go

@@ -470,7 +470,7 @@ func (c *Client) Status() (peer.FullStatus, error) {
 	if connect != nil {
 		engine := connect.Engine()
 		if engine != nil {
-			_ = engine.RunHealthProbes(context.Background(), false)
+			_ = engine.RunHealthProbes(false)
 		}
 	}
 

client/firewall/uspfilter/filter.go

@@ -121,7 +121,6 @@ type Manager struct {
 	udpTracker     *conntrack.UDPTracker
 	icmpTracker    *conntrack.ICMPTracker
 	tcpTracker     *conntrack.TCPTracker
-	fragments      *fragmentTracker
 	forwarder      atomic.Pointer[forwarder.Forwarder]
 	pendingCapture atomic.Pointer[forwarder.PacketCapture]
 	logger         *nblog.Logger
@@ -184,41 +183,6 @@ func (d *decoder) decodePacket(data []byte) error {
 	}
 }
 
-// decodeTransport decodes the transport header of a first fragment (which
-// gopacket leaves undecoded) into the decoder and appends its layer type to
-// decoded, so the ACL pipeline can evaluate it like a normal packet. It returns
-// false if the protocol is unsupported or the header is truncated.
-func (d *decoder) decodeTransport(proto layers.IPProtocol, payload []byte) bool {
-	var l4 gopacket.DecodingLayer
-	var layerType gopacket.LayerType
-	var minLen int
-	switch proto {
-	case layers.IPProtocolTCP:
-		l4, layerType, minLen = &d.tcp, layers.LayerTypeTCP, 20
-	case layers.IPProtocolUDP:
-		l4, layerType, minLen = &d.udp, layers.LayerTypeUDP, 8
-	case layers.IPProtocolICMPv4:
-		l4, layerType, minLen = &d.icmp4, layers.LayerTypeICMPv4, 8
-	case layers.IPProtocolICMPv6:
-		l4, layerType, minLen = &d.icmp6, layers.LayerTypeICMPv6, 8
-	default:
-		return false
-	}
-
-	// Reject a fragment too small to hold the full transport header before
-	// decoding: it can't be ACL-evaluated (tiny-fragment attack), and skipping
-	// the decode avoids gopacket allocating an error on the drop path.
-	if len(payload) < minLen {
-		return false
-	}
-
-	if err := l4.DecodeFromBytes(payload, gopacket.NilDecodeFeedback); err != nil {
-		return false
-	}
-	d.decoded = append(d.decoded, layerType)
-	return true
-}
-
 // Create userspace firewall manager constructor
 func Create(iface common.IFaceMapper, disableServerRoutes bool, flowLogger nftypes.FlowLogger, mtu uint16) (*Manager, error) {
 	return create(iface, nil, disableServerRoutes, flowLogger, mtu)
@@ -322,8 +286,6 @@ func create(iface common.IFaceMapper, nativeFirewall firewall.Manager, disableSe
 	if err := m.localipmanager.UpdateLocalIPs(iface); err != nil {
 		return nil, fmt.Errorf("update local IPs: %w", err)
 	}
-	m.fragments = newFragmentTracker(m.logger)
-
 	if disableConntrack {
 		log.Info("conntrack is disabled")
 	} else {
@@ -337,7 +299,6 @@ func create(iface common.IFaceMapper, nativeFirewall firewall.Manager, disableSe
 		}
 	}
 	if err := iface.SetFilter(m); err != nil {
-		m.fragments.Close()
 		return nil, fmt.Errorf("set filter: %w", err)
 	}
 	return m, nil
@@ -733,10 +694,6 @@ func (m *Manager) resetState() {
 		m.tcpTracker.Close()
 	}
 
-	if m.fragments != nil {
-		m.fragments.Close()
-	}
-
 	if fwder := m.forwarder.Load(); fwder != nil {
 		fwder.SetCapture(nil)
 		fwder.Stop()
@@ -1089,20 +1046,19 @@ func (m *Manager) filterInbound(packetData []byte, size int) bool {
 		return true
 	}
 
-	// gopacket does not decode the transport header of any IP fragment, so
+	// TODO: pass fragments of routed packets to forwarder
-	// fragments take a dedicated path: the first fragment's header is decoded
-	// and ACL-evaluated here, and the remaining fragments inherit its verdict.
 	if fragment {
-		return m.filterInboundFragment(d, srcIP, dstIP, size)
+		if m.logger.Enabled(nblog.LevelTrace) {
+			if d.decoded[0] == layers.LayerTypeIPv4 {
+				m.logger.Trace4("packet is a fragment: src=%v dst=%v id=%v flags=%v",
+					srcIP, dstIP, d.ip4.Id, d.ip4.Flags)
+			} else {
+				m.logger.Trace2("packet is an IPv6 fragment: src=%v dst=%v", srcIP, dstIP)
+			}
+		}
+		return false
 	}
 
-	return m.filterInboundDecoded(d, srcIP, dstIP, packetData, size)
-}
-
-// filterInboundDecoded runs the ACL, DNAT and conntrack pipeline on a fully
-// decoded (non-fragment) inbound packet. It returns true if the packet should
-// be dropped.
-func (m *Manager) filterInboundDecoded(d *decoder, srcIP, dstIP netip.Addr, packetData []byte, size int) bool {
 	// TODO: optimize port DNAT by caching matched rules in conntrack
 	if translated := m.translateInboundPortDNAT(packetData, d, srcIP, dstIP); translated {
 		// Re-decode after port DNAT translation to update port information
@@ -1133,226 +1089,33 @@ func (m *Manager) filterInboundDecoded(d *decoder, srcIP, dstIP netip.Addr, pack
 	return m.handleRoutedTraffic(d, srcIP, dstIP, packetData, size)
 }
 
-// fragmentMeta holds the reassembly identity and layout of an IP fragment,
-// extracted uniformly for IPv4 and IPv6.
-type fragmentMeta struct {
-	key fragmentKey
-	// offset is the fragment offset in 8-byte units (zero for the first
-	// fragment).
-	offset uint16
-	// moreFragments is the More Fragments bit. A first fragment with it unset is
-	// an IPv6 atomic fragment (a complete datagram, RFC 6946): it has no trailing
-	// fragments to inherit a verdict, so it must not be recorded.
-	moreFragments bool
-	proto         layers.IPProtocol
-	// l4payload is the fragmentable payload of this fragment. For the first
-	// fragment it starts with the transport header.
-	l4payload []byte
-	// headerEndOctets is the first fragment's payload length in 8-byte units:
-	// the smallest offset a trailing fragment may start at without overlapping
-	// the inspected transport header.
-	headerEndOctets uint16
-}
-
-// fragmentMetadata extracts the fragment identity and layout from a decoded IP
-// fragment. It returns false for fragments it can't interpret (e.g. an IPv6
-// fragment header shorter than 8 bytes), which are then dropped.
-func fragmentMetadata(d *decoder, srcIP, dstIP netip.Addr) (fragmentMeta, bool) {
-	switch d.decoded[0] {
-	case layers.LayerTypeIPv4:
-		payload := d.ip4.Payload
-		return fragmentMeta{
-			key:             fragmentKey{srcIP: srcIP, dstIP: dstIP, id: uint32(d.ip4.Id), proto: uint8(d.ip4.Protocol)},
-			offset:          d.ip4.FragOffset,
-			moreFragments:   d.ip4.Flags&layers.IPv4MoreFragments != 0,
-			proto:           d.ip4.Protocol,
-			l4payload:       payload,
-			headerEndOctets: octets(len(payload)),
-		}, true
-
-	case layers.LayerTypeIPv6:
-		// IPv6 fragment extension header: 8 bytes, followed by the fragmentable
-		// payload. Layout: next header (1), reserved (1), offset+flags (2), id (4).
-		payload := d.ip6.Payload
-		if len(payload) < 8 {
-			return fragmentMeta{}, false
-		}
-		nextHeader := layers.IPProtocol(payload[0])
-		offsetFlags := binary.BigEndian.Uint16(payload[2:4])
-		id := binary.BigEndian.Uint32(payload[4:8])
-		l4 := payload[8:]
-		return fragmentMeta{
-			key:             fragmentKey{srcIP: srcIP, dstIP: dstIP, id: id, proto: uint8(nextHeader)},
-			offset:          offsetFlags >> 3,
-			moreFragments:   offsetFlags&1 != 0,
-			proto:           nextHeader,
-			l4payload:       l4,
-			headerEndOctets: octets(len(l4)),
-		}, true
-
-	default:
-		return fragmentMeta{}, false
-	}
-}
-
-// octets rounds a byte length up to whole 8-byte units, the granularity of the
-// IP fragment offset field.
-func octets(nbytes int) uint16 {
-	return uint16((nbytes + 7) / 8)
-}
-
-// filterInboundFragment decides the fate of an IP fragment. gopacket stops
-// decoding at the network layer for every fragment, so the first fragment's
-// transport header is decoded and ACL-evaluated here and its verdict recorded;
-// the remaining (headerless) fragments inherit that verdict. Anything that
-// cannot be tied to an allowed, non-overlapping first fragment is dropped.
-func (m *Manager) filterInboundFragment(d *decoder, srcIP, dstIP netip.Addr, size int) bool {
-	meta, ok := fragmentMetadata(d, srcIP, dstIP)
-	if !ok {
-		if m.logger.Enabled(nblog.LevelTrace) {
-			m.logger.Trace2("dropping unsupported fragment: src=%v dst=%v", srcIP, dstIP)
-		}
-		return true
-	}
-
-	if meta.offset != 0 {
-		return m.filterTrailingFragment(meta, srcIP, dstIP)
-	}
-
-	// A new first fragment supersedes any recorded verdict for this datagram, so
-	// a re-sent or overlapping offset-zero fragment can't inherit the old one.
-	m.fragments.poison(meta.key)
-
-	// First fragment: decode its transport header so the ACL can evaluate it. A
-	// decode failure means the fragment is too small to hold the full transport
-	// header (RFC 1858 §3 tiny-fragment attack); it can't be evaluated, so drop it.
-	if !d.decodeTransport(meta.proto, meta.l4payload) {
-		if m.logger.Enabled(nblog.LevelTrace) {
-			m.logger.Trace3("dropping first fragment without full L4 header: src=%v dst=%v id=%v",
-				srcIP, dstIP, meta.key.id)
-		}
-		return true
-	}
-
-	return m.filterFirstFragment(d, meta, srcIP, dstIP, size)
-}
-
-// filterTrailingFragment applies a recorded first-fragment verdict to a
-// non-first fragment.
-func (m *Manager) filterTrailingFragment(meta fragmentMeta, srcIP, dstIP netip.Addr) bool {
-	switch m.fragments.verdict(meta.key, meta.offset) {
-	case fragmentAllow:
-		return false
-	case fragmentOverlap:
-		if m.logger.Enabled(nblog.LevelTrace) {
-			m.logger.Trace3("dropping overlapping fragment rewriting inspected header: src=%v dst=%v id=%v",
-				srcIP, dstIP, meta.key.id)
-		}
-		return true
-	default:
-		if m.logger.Enabled(nblog.LevelTrace) {
-			m.logger.Trace3("dropping fragment with no allowed first fragment: src=%v dst=%v id=%v",
-				srcIP, dstIP, meta.key.id)
-		}
-		return true
-	}
-}
-
-// filterFirstFragment runs the verdict part of the inbound pipeline on a first
-// fragment with its transport header decoded. It mirrors filterInboundDecoded
-// but skips DNAT (port rewriting on fragments is unsupported) and forwarder
-// injection (fragments are left to the stack to reassemble, not forwarded).
-// Allowed fragments have their verdict recorded so the datagram's trailing
-// fragments inherit it.
-func (m *Manager) filterFirstFragment(d *decoder, meta fragmentMeta, srcIP, dstIP netip.Addr, size int) bool {
-	if m.stateful && m.isValidTrackedConnection(d, srcIP, dstIP, size) {
-		m.recordFirstFragment(meta)
-		return false
-	}
-
-	if m.localipmanager.IsLocalIP(dstIP) {
-		ruleID, blocked := m.peerACLsBlock(srcIP, d, nil)
-		if blocked {
-			m.storeDropFlow("Dropping local first fragment (ACL denied): rule_id=%s proto=%v src=%s:%d dst=%s:%d",
-				d, srcIP, dstIP, ruleID, size)
-			return true
-		}
-		m.trackInbound(d, srcIP, dstIP, ruleID, size)
-		m.recordFirstFragment(meta)
-		return false
-	}
-
-	if !m.routingEnabled.Load() {
-		if m.logger.Enabled(nblog.LevelTrace) {
-			m.logger.Trace2("Dropping routed fragment (routing disabled): src=%s dst=%s", srcIP, dstIP)
-		}
-		return true
-	}
-	if m.nativeRouter.Load() {
-		m.trackInbound(d, srcIP, dstIP, nil, size)
-		m.recordFirstFragment(meta)
-		return false
-	}
-
-	// TODO: pass fragments of routed packets to the forwarder; until then
-	// allowed routed fragments go to the native stack.
-	srcPort, dstPort := getPortsFromPacket(d)
-	ruleID, pass := m.routeACLsPass(srcIP, dstIP, d.decoded[1], srcPort, dstPort)
-	if !pass {
-		m.storeDropFlow("Dropping routed first fragment (ACL denied): rule_id=%s proto=%v src=%s:%d dst=%s:%d",
-			d, srcIP, dstIP, ruleID, size)
-		return true
-	}
-
-	m.recordFirstFragment(meta)
-	return false
-}
-
-// recordFirstFragment caches an allowed first fragment's verdict for its
-// trailing fragments to inherit. Atomic fragments (no More Fragments bit) are
-// complete datagrams with no trailing fragments, so they are not cached and
-// cannot exhaust the verdict table.
-func (m *Manager) recordFirstFragment(meta fragmentMeta) {
-	if !meta.moreFragments {
-		return
-	}
-	m.fragments.recordAllowed(meta.key, meta.headerEndOctets)
-}
-
-// storeDropFlow logs and records a netflow drop event for an inbound packet
-// denied by the ACLs. msg is the trace format taking rule id, protocol, source
-// and destination.
-func (m *Manager) storeDropFlow(msg string, d *decoder, srcIP, dstIP netip.Addr, ruleID []byte, size int) {
-	pnum := getProtocolFromPacket(d)
-	srcPort, dstPort := getPortsFromPacket(d)
-
-	if m.logger.Enabled(nblog.LevelTrace) {
-		m.logger.Trace6(msg, ruleID, pnum, srcIP, srcPort, dstIP, dstPort)
-	}
-
-	m.flowLogger.StoreEvent(nftypes.EventFields{
-		FlowID:     uuid.New(),
-		Type:       nftypes.TypeDrop,
-		RuleID:     ruleID,
-		Direction:  nftypes.Ingress,
-		Protocol:   pnum,
-		SourceIP:   srcIP,
-		DestIP:     dstIP,
-		SourcePort: srcPort,
-		DestPort:   dstPort,
-		// TODO: icmp type/code
-		RxPackets: 1,
-		RxBytes:   uint64(size),
-	})
-}
-
 // handleLocalTraffic handles local traffic.
 // If it returns true, the packet should be dropped.
 func (m *Manager) handleLocalTraffic(d *decoder, srcIP, dstIP netip.Addr, packetData []byte, size int) bool {
 	ruleID, blocked := m.peerACLsBlock(srcIP, d, packetData)
 	if blocked {
-		m.storeDropFlow("Dropping local packet (ACL denied): rule_id=%s proto=%v src=%s:%d dst=%s:%d",
+		pnum := getProtocolFromPacket(d)
-			d, srcIP, dstIP, ruleID, size)
+		srcPort, dstPort := getPortsFromPacket(d)
+
+		if m.logger.Enabled(nblog.LevelTrace) {
+			m.logger.Trace6("Dropping local packet (ACL denied): rule_id=%s proto=%v src=%s:%d dst=%s:%d",
+				ruleID, pnum, srcIP, srcPort, dstIP, dstPort)
+		}
+
+		m.flowLogger.StoreEvent(nftypes.EventFields{
+			FlowID:     uuid.New(),
+			Type:       nftypes.TypeDrop,
+			RuleID:     ruleID,
+			Direction:  nftypes.Ingress,
+			Protocol:   pnum,
+			SourceIP:   srcIP,
+			DestIP:     dstIP,
+			SourcePort: srcPort,
+			DestPort:   dstPort,
+			// TODO: icmp type/code
+			RxPackets: 1,
+			RxBytes:   uint64(size),
+		})
 		return true
 	}
 
@@ -1405,8 +1168,27 @@ func (m *Manager) handleRoutedTraffic(d *decoder, srcIP, dstIP netip.Addr, packe
 
 	ruleID, pass := m.routeACLsPass(srcIP, dstIP, protoLayer, srcPort, dstPort)
 	if !pass {
-		m.storeDropFlow("Dropping routed packet (ACL denied): rule_id=%s proto=%v src=%s:%d dst=%s:%d",
+		proto := getProtocolFromPacket(d)
-			d, srcIP, dstIP, ruleID, size)
+
+		if m.logger.Enabled(nblog.LevelTrace) {
+			m.logger.Trace6("Dropping routed packet (ACL denied): rule_id=%s proto=%v src=%s:%d dst=%s:%d",
+				ruleID, proto, srcIP, srcPort, dstIP, dstPort)
+		}
+
+		m.flowLogger.StoreEvent(nftypes.EventFields{
+			FlowID:     uuid.New(),
+			Type:       nftypes.TypeDrop,
+			RuleID:     ruleID,
+			Direction:  nftypes.Ingress,
+			Protocol:   proto,
+			SourceIP:   srcIP,
+			DestIP:     dstIP,
+			SourcePort: srcPort,
+			DestPort:   dstPort,
+			// TODO: icmp type/code
+			RxPackets: 1,
+			RxBytes:   uint64(size),
+		})
 		return true
 	}
 

client/firewall/uspfilter/forwarder/forwarder.go

@@ -5,9 +5,7 @@ import (
 	"fmt"
 	"net"
 	"net/netip"
-	"os"
 	"runtime"
-	"strconv"
 	"sync"
 	"time"
 
@@ -33,11 +31,6 @@ const (
 	defaultMaxInFlight   = 1024
 	iosReceiveWindow     = 16384
 	iosMaxInFlight       = 256
-
-	// envForceTCPRACK overrides the platform default for gVisor's RACK loss
-	// detection. Set to a truthy value to force RACK on, or a falsy value to
-	// force it off, on any platform.
-	envForceTCPRACK = "NB_FORCE_TCP_RACK"
 )
 
 type Forwarder struct {
@@ -159,8 +152,6 @@ func New(iface common.IFaceMapper, logger *nblog.Logger, flowLogger nftypes.Flow
 		maxInFlight = iosMaxInFlight
 	}
 
-	configureTCPRecovery(s)
-
 	tcpForwarder := tcp.NewForwarder(s, receiveWindow, maxInFlight, f.handleTCP)
 	s.SetTransportProtocolHandler(tcp.ProtocolNumber, tcpForwarder.HandlePacket)
 
@@ -475,31 +466,3 @@ func probeRawICMP(network, addr string, logger *nblog.Logger) bool {
 	logger.Debug1("forwarder: raw %s socket access available", network)
 	return true
 }
-
-// configureTCPRecovery disables gVisor's RACK loss detection on Windows, where
-// it interacts poorly with the host and collapses throughput on routed TCP
-// connections (gVisor issue #9778). Other platforms keep the default. The
-// EnvForceTCPRACK environment variable overrides the platform default.
-func configureTCPRecovery(s *stack.Stack) {
-	disableRACK := runtime.GOOS == "windows"
-
-	if val := os.Getenv(envForceTCPRACK); val != "" {
-		force, err := strconv.ParseBool(val)
-		if err != nil {
-			log.Warnf("parse %s: %v", envForceTCPRACK, err)
-		} else {
-			disableRACK = !force
-		}
-	}
-
-	if !disableRACK {
-		return
-	}
-
-	opt := tcpip.TCPRecovery(0)
-	if err := s.SetTransportProtocolOption(tcp.ProtocolNumber, &opt); err != nil {
-		log.Warnf("disable TCP RACK loss detection: %v", err)
-		return
-	}
-	log.Info("forwarder: TCP RACK loss detection disabled")
-}

client/firewall/uspfilter/fragment.go

@@ -1,204 +0,0 @@
-package uspfilter
-
-import (
-	"context"
-	"net/netip"
-	"os"
-	"strconv"
-	"sync"
-	"time"
-
-	nblog "github.com/netbirdio/netbird/client/firewall/uspfilter/log"
-)
-
-const (
-	// defaultFragmentTimeout bounds how long a first-fragment verdict is kept
-	// while the remaining fragments arrive. It mirrors the Linux IP reassembly
-	// timeout (net.ipv4.ipfrag_time).
-	defaultFragmentTimeout = 30 * time.Second
-	// fragmentCleanupInterval is how often expired verdicts are purged.
-	fragmentCleanupInterval = 10 * time.Second
-	// defaultMaxFragmentEntries caps the number of concurrently tracked
-	// fragmented datagrams. The table stays bounded because each datagram is a
-	// single small entry regardless of how many fragments it is split into, and
-	// the 13-bit IPv4 fragment-offset field limits any datagram to 64 KiB.
-	defaultMaxFragmentEntries = 16384
-
-	// EnvFragmentMaxEntries overrides defaultMaxFragmentEntries.
-	EnvFragmentMaxEntries = "NB_FRAGMENT_MAX_ENTRIES"
-)
-
-// fragmentVerdict is the decision for a trailing (headerless) fragment.
-type fragmentVerdict int
-
-const (
-	// fragmentDeny drops the fragment: no allowed first fragment is on record.
-	fragmentDeny fragmentVerdict = iota
-	// fragmentAllow passes the fragment: it belongs to an allowed datagram and
-	// does not overlap the already-inspected transport header.
-	fragmentAllow
-	// fragmentOverlap drops the fragment and poisons its datagram: it overlaps
-	// the transport header the ACL inspected (RFC 1858 §4, RFC 3128; RFC 5722
-	// requires discarding the whole datagram on overlap for IPv6).
-	fragmentOverlap
-)
-
-// fragmentKey identifies a fragmented datagram. It matches the RFC 791 / RFC
-// 8200 reassembly key: source, destination, protocol and identification. The id
-// is 32-bit to hold both the IPv4 (16-bit) and IPv6 (32-bit) identification.
-type fragmentKey struct {
-	srcIP netip.Addr
-	dstIP netip.Addr
-	id    uint32
-	proto uint8
-}
-
-// fragmentEntry records the verdict of an allowed first fragment.
-type fragmentEntry struct {
-	// headerEndOctets is the offset, in 8-byte units, at which the first
-	// fragment's payload ended. A trailing fragment starting before this
-	// overlaps bytes the ACL already inspected and is rejected.
-	headerEndOctets uint16
-	// recordedAt is when the first fragment was accepted. The verdict expires a
-	// fixed timeout later and is not refreshed, mirroring the kernel reassembly
-	// timer so a trailing-fragment flood can't keep a datagram alive.
-	recordedAt time.Time
-}
-
-// fragmentTracker records the ACL verdict of a datagram's first fragment so the
-// remaining fragments, which carry no L4 header, can inherit the decision
-// without reassembling the datagram. Only allowed first fragments are stored;
-// anything that cannot be tied to an allowed, non-overlapping first fragment is
-// dropped (fail closed).
-type fragmentTracker struct {
-	logger  *nblog.Logger
-	mutex   sync.Mutex
-	entries map[fragmentKey]fragmentEntry
-	timeout time.Duration
-	// maxEntries caps the table; atCapacity dedups the capacity warning until
-	// the table drains below the cap again.
-	maxEntries    int
-	atCapacity    bool
-	cleanupTicker *time.Ticker
-	cancel        context.CancelFunc
-}
-
-func newFragmentTracker(logger *nblog.Logger) *fragmentTracker {
-	ctx, cancel := context.WithCancel(context.Background())
-	t := &fragmentTracker{
-		logger:        logger,
-		entries:       make(map[fragmentKey]fragmentEntry),
-		timeout:       defaultFragmentTimeout,
-		maxEntries:    fragmentMaxEntries(logger),
-		cleanupTicker: time.NewTicker(fragmentCleanupInterval),
-		cancel:        cancel,
-	}
-	go t.cleanupRoutine(ctx)
-	return t
-}
-
-func fragmentMaxEntries(logger *nblog.Logger) int {
-	v := os.Getenv(EnvFragmentMaxEntries)
-	if v == "" {
-		return defaultMaxFragmentEntries
-	}
-	n, err := strconv.Atoi(v)
-	if err != nil || n <= 0 {
-		logger.Warn2("invalid %s=%q, using default", EnvFragmentMaxEntries, v)
-		return defaultMaxFragmentEntries
-	}
-	return n
-}
-
-// recordAllowed stores the verdict of an allowed first fragment. headerEndOctets
-// is the first fragment's payload length in 8-byte units. When the table is full
-// the record is dropped, which fails closed: the datagram's trailing fragments
-// will be denied.
-func (t *fragmentTracker) recordAllowed(key fragmentKey, headerEndOctets uint16) {
-	t.mutex.Lock()
-	defer t.mutex.Unlock()
-
-	if t.entries == nil {
-		return
-	}
-	if _, ok := t.entries[key]; !ok && len(t.entries) >= t.maxEntries {
-		if !t.atCapacity {
-			t.atCapacity = true
-			t.logger.Warn2("fragment verdict table at capacity (%d/%d): trailing fragments of new datagrams will be dropped",
-				len(t.entries), t.maxEntries)
-		}
-		return
-	}
-	t.entries[key] = fragmentEntry{
-		headerEndOctets: headerEndOctets,
-		recordedAt:      time.Now(),
-	}
-}
-
-// poison drops any recorded verdict for a datagram, so its later fragments are
-// denied until a new allowed first fragment is recorded. Called on every
-// offset-zero fragment to defeat offset-zero overlap rewrites (RFC 3128).
-func (t *fragmentTracker) poison(key fragmentKey) {
-	t.mutex.Lock()
-	defer t.mutex.Unlock()
-	delete(t.entries, key)
-}
-
-// verdict decides the fate of a trailing fragment at fragOffsetOctets (the IPv4
-// fragment offset, in 8-byte units). A fragment overlapping the inspected
-// header poisons the datagram: the entry is removed so all further fragments of
-// that datagram are denied too.
-func (t *fragmentTracker) verdict(key fragmentKey, fragOffsetOctets uint16) fragmentVerdict {
-	t.mutex.Lock()
-	defer t.mutex.Unlock()
-
-	entry, ok := t.entries[key]
-	if !ok {
-		return fragmentDeny
-	}
-	if time.Since(entry.recordedAt) > t.timeout {
-		delete(t.entries, key)
-		return fragmentDeny
-	}
-	if fragOffsetOctets < entry.headerEndOctets {
-		delete(t.entries, key)
-		return fragmentOverlap
-	}
-	return fragmentAllow
-}
-
-func (t *fragmentTracker) cleanupRoutine(ctx context.Context) {
-	defer t.cleanupTicker.Stop()
-	for {
-		select {
-		case <-t.cleanupTicker.C:
-			t.cleanup()
-		case <-ctx.Done():
-			return
-		}
-	}
-}
-
-func (t *fragmentTracker) cleanup() {
-	t.mutex.Lock()
-	defer t.mutex.Unlock()
-
-	for key, entry := range t.entries {
-		if time.Since(entry.recordedAt) > t.timeout {
-			delete(t.entries, key)
-		}
-	}
-
-	if len(t.entries) < t.maxEntries {
-		t.atCapacity = false
-	}
-}
-
-// Close stops the cleanup routine and releases resources.
-func (t *fragmentTracker) Close() {
-	t.cancel()
-
-	t.mutex.Lock()
-	t.entries = nil
-	t.mutex.Unlock()
-}

client/firewall/uspfilter/fragment_bench_test.go

@@ -1,115 +0,0 @@
-package uspfilter
-
-import (
-	"encoding/binary"
-	"testing"
-)
-
-// benchFilterInbound drives filterInbound over a fixed packet in a tight loop.
-// Packets are built once, outside the timed region, so the benchmark measures
-// only pipeline cost, which is what an attacker can amplify.
-func benchFilterInbound(b *testing.B, pkt []byte) {
-	b.Helper()
-	b.ReportAllocs()
-	b.SetBytes(int64(len(pkt)))
-	b.ResetTimer()
-	for i := 0; i < b.N; i++ {
-		m := benchManager
-		m.filterInbound(pkt, len(pkt))
-	}
-}
-
-// benchManager is a package-level manager reused across fragment benchmarks so
-// setup cost stays out of the timed region.
-var benchManager *Manager
-
-func setupBenchManager(b *testing.B) *Manager {
-	b.Helper()
-	m := newFragmentTestManager(b)
-	allowUDP(b, m, 8080)
-	// Disable conntrack so the allowed-first-fragment path measures transport
-	// decode + ACL every iteration instead of matching the connection tracked
-	// on the first iteration.
-	m.stateful = false
-	benchManager = m
-	return m
-}
-
-// BenchmarkInbound_NormalPacket is the baseline: a full, non-fragmented UDP
-// packet that passes the ACL. Fragment paths should stay comparable to this.
-func BenchmarkInbound_NormalPacket(b *testing.B) {
-	setupBenchManager(b)
-	pkt := normalUDPPacket(b, 8080, 32)
-	benchFilterInbound(b, pkt)
-}
-
-// BenchmarkInbound_FirstFragmentAllowed measures the first-fragment path:
-// transport decode + ACL evaluation + verdict record.
-func BenchmarkInbound_FirstFragmentAllowed(b *testing.B) {
-	setupBenchManager(b)
-	pkt := firstFragmentUDP(b, 0x2000, 8080, 32)
-	benchFilterInbound(b, pkt)
-}
-
-// BenchmarkInbound_TrailingFragmentAllowed measures the common trailing-fragment
-// path: a single map lookup after the first fragment is on record.
-func BenchmarkInbound_TrailingFragmentAllowed(b *testing.B) {
-	m := setupBenchManager(b)
-	first := firstFragmentUDP(b, 0x3000, 8080, 32)
-	m.filterInbound(first, len(first))
-	pkt := trailingFragment(b, 0x3000, 5, false, 24)
-	benchFilterInbound(b, pkt)
-}
-
-// BenchmarkInbound_TrailingFragmentNoFirst is the primary DoS vector: an
-// attacker floods trailing fragments with no first fragment on record. Each is
-// a map miss and must be cheap.
-func BenchmarkInbound_TrailingFragmentNoFirst(b *testing.B) {
-	setupBenchManager(b)
-	pkt := trailingFragment(b, 0x4000, 185, false, 40)
-	benchFilterInbound(b, pkt)
-}
-
-// BenchmarkInbound_TinyFirstFragment measures the tiny-fragment drop path: a
-// first fragment too small to decode a transport header.
-func BenchmarkInbound_TinyFirstFragment(b *testing.B) {
-	setupBenchManager(b)
-	pkt := trailingFragment(b, 0x5000, 0, true, 4)
-	benchFilterInbound(b, pkt)
-}
-
-// BenchmarkInbound_TrailingFragmentDistinctIDs is the worst case for the
-// verdict table: an attacker varies the datagram id on every packet so no first
-// fragment ever matches. Verdict lookups always miss and nothing is recorded,
-// so the table cannot grow. Each iteration rewrites the id field in place.
-func BenchmarkInbound_TrailingFragmentDistinctIDs(b *testing.B) {
-	setupBenchManager(b)
-	pkt := trailingFragment(b, 0x6000, 185, false, 40)
-	m := benchManager
-
-	b.ReportAllocs()
-	b.SetBytes(int64(len(pkt)))
-	b.ResetTimer()
-	for i := 0; i < b.N; i++ {
-		// IPv4 identification field is at bytes 4:6.
-		binary.BigEndian.PutUint16(pkt[4:6], uint16(i))
-		m.filterInbound(pkt, len(pkt))
-	}
-}
-
-// BenchmarkInbound_FirstFragmentDistinctIDs measures sustained first-fragment
-// pressure with distinct ids: transport decode + ACL + verdict insert until the
-// table caps, exercising the map growth and capacity guard.
-func BenchmarkInbound_FirstFragmentDistinctIDs(b *testing.B) {
-	setupBenchManager(b)
-	pkt := firstFragmentUDP(b, 0x7000, 8080, 32)
-	m := benchManager
-
-	b.ReportAllocs()
-	b.SetBytes(int64(len(pkt)))
-	b.ResetTimer()
-	for i := 0; i < b.N; i++ {
-		binary.BigEndian.PutUint16(pkt[4:6], uint16(i))
-		m.filterInbound(pkt, len(pkt))
-	}
-}

client/firewall/uspfilter/fragment_test.go

@@ -1,554 +0,0 @@
-package uspfilter
-
-import (
-	"encoding/binary"
-	"net"
-	"net/netip"
-	"testing"
-	"time"
-
-	"github.com/google/gopacket"
-	"github.com/google/gopacket/layers"
-	"github.com/stretchr/testify/require"
-
-	fw "github.com/netbirdio/netbird/client/firewall/manager"
-	nbiface "github.com/netbirdio/netbird/client/iface"
-	"github.com/netbirdio/netbird/client/iface/device"
-	"github.com/netbirdio/netbird/client/iface/wgaddr"
-)
-
-const (
-	fragTestSrc   = "100.10.0.1"
-	fragTestDst   = "100.10.0.100"
-	fragTestSrcV6 = "fd00::1"
-	fragTestDstV6 = "fd00::100"
-)
-
-func newFragmentTestManager(tb testing.TB) *Manager {
-	tb.Helper()
-
-	ifaceMock := &IFaceMock{
-		SetFilterFunc: func(device.PacketFilter) error { return nil },
-		AddressFunc: func() wgaddr.Address {
-			return wgaddr.Address{
-				IP:      netip.MustParseAddr(fragTestDst),
-				Network: netip.MustParsePrefix("100.10.0.0/16"),
-				IPv6:    netip.MustParseAddr(fragTestDstV6),
-				IPv6Net: netip.MustParsePrefix("fd00::/64"),
-			}
-		},
-	}
-
-	m, err := Create(ifaceMock, false, flowLogger, nbiface.DefaultMTU)
-	require.NoError(tb, err)
-	require.NoError(tb, m.UpdateLocalIPs())
-	tb.Cleanup(func() { require.NoError(tb, m.Close(nil)) })
-	return m
-}
-
-// firstFragmentUDPTo builds the first fragment of a fragmented UDP datagram to
-// the given destination: it carries the full UDP header plus payloadLen bytes
-// of data, with the More Fragments flag set and offset zero.
-func firstFragmentUDPTo(tb testing.TB, dst string, id uint16, dstPort uint16, payloadLen int) []byte {
-	tb.Helper()
-
-	ip := &layers.IPv4{
-		Version:  4,
-		TTL:      64,
-		Id:       id,
-		Protocol: layers.IPProtocolUDP,
-		SrcIP:    net.ParseIP(fragTestSrc),
-		DstIP:    net.ParseIP(dst),
-		Flags:    layers.IPv4MoreFragments,
-	}
-	udp := &layers.UDP{SrcPort: 40000, DstPort: layers.UDPPort(dstPort)}
-	require.NoError(tb, udp.SetNetworkLayerForChecksum(ip))
-
-	buf := gopacket.NewSerializeBuffer()
-	opts := gopacket.SerializeOptions{ComputeChecksums: true, FixLengths: true}
-	require.NoError(tb, gopacket.SerializeLayers(buf, opts, ip, udp, gopacket.Payload(make([]byte, payloadLen))))
-	return buf.Bytes()
-}
-
-func firstFragmentUDP(tb testing.TB, id uint16, dstPort uint16, payloadLen int) []byte {
-	tb.Helper()
-	return firstFragmentUDPTo(tb, fragTestDst, id, dstPort, payloadLen)
-}
-
-// firstFragmentTCP builds the first fragment of a fragmented TCP datagram: the
-// full 20-byte TCP header plus 12 bytes of data, with the More Fragments flag
-// set and offset zero.
-func firstFragmentTCP(tb testing.TB, id uint16, dstPort uint16) []byte {
-	tb.Helper()
-
-	ip := &layers.IPv4{
-		Version:  4,
-		TTL:      64,
-		Id:       id,
-		Protocol: layers.IPProtocolTCP,
-		SrcIP:    net.ParseIP(fragTestSrc),
-		DstIP:    net.ParseIP(fragTestDst),
-		Flags:    layers.IPv4MoreFragments,
-	}
-	tcp := &layers.TCP{SrcPort: 40000, DstPort: layers.TCPPort(dstPort), SYN: true, Window: 64240}
-	require.NoError(tb, tcp.SetNetworkLayerForChecksum(ip))
-
-	buf := gopacket.NewSerializeBuffer()
-	opts := gopacket.SerializeOptions{ComputeChecksums: true, FixLengths: true}
-	require.NoError(tb, gopacket.SerializeLayers(buf, opts, ip, tcp, gopacket.Payload(make([]byte, 12))))
-	return buf.Bytes()
-}
-
-// trailingFragmentTo builds a non-first fragment to the given destination: an
-// IPv4 header at the given fragment offset (in 8-byte units) carrying raw
-// payload and no L4 header.
-func trailingFragmentTo(tb testing.TB, dst string, proto layers.IPProtocol, id uint16, fragOffsetOctets uint16, moreFragments bool, payloadLen int) []byte {
-	tb.Helper()
-
-	ip := &layers.IPv4{
-		Version:    4,
-		TTL:        64,
-		Id:         id,
-		Protocol:   proto,
-		SrcIP:      net.ParseIP(fragTestSrc),
-		DstIP:      net.ParseIP(dst),
-		FragOffset: fragOffsetOctets,
-	}
-	if moreFragments {
-		ip.Flags = layers.IPv4MoreFragments
-	}
-
-	buf := gopacket.NewSerializeBuffer()
-	opts := gopacket.SerializeOptions{FixLengths: true}
-	require.NoError(tb, gopacket.SerializeLayers(buf, opts, ip, gopacket.Payload(make([]byte, payloadLen))))
-	return buf.Bytes()
-}
-
-func trailingFragment(tb testing.TB, id uint16, fragOffsetOctets uint16, moreFragments bool, payloadLen int) []byte {
-	tb.Helper()
-	return trailingFragmentTo(tb, fragTestDst, layers.IPProtocolUDP, id, fragOffsetOctets, moreFragments, payloadLen)
-}
-
-// outboundUDPPacket builds a complete outbound UDP packet from the local
-// address, used to establish conntrack state for reply-direction tests.
-func outboundUDPPacket(tb testing.TB, srcPort, dstPort uint16) []byte {
-	tb.Helper()
-
-	ip := &layers.IPv4{
-		Version:  4,
-		TTL:      64,
-		Id:       1,
-		Protocol: layers.IPProtocolUDP,
-		SrcIP:    net.ParseIP(fragTestDst),
-		DstIP:    net.ParseIP(fragTestSrc),
-	}
-	udp := &layers.UDP{SrcPort: layers.UDPPort(srcPort), DstPort: layers.UDPPort(dstPort)}
-	require.NoError(tb, udp.SetNetworkLayerForChecksum(ip))
-
-	buf := gopacket.NewSerializeBuffer()
-	opts := gopacket.SerializeOptions{ComputeChecksums: true, FixLengths: true}
-	require.NoError(tb, gopacket.SerializeLayers(buf, opts, ip, udp, gopacket.Payload(make([]byte, 16))))
-	return buf.Bytes()
-}
-
-// normalUDPPacket builds a complete, non-fragmented UDP packet for baseline
-// comparisons against the fragment paths.
-func normalUDPPacket(tb testing.TB, dstPort uint16, payloadLen int) []byte {
-	tb.Helper()
-
-	ip := &layers.IPv4{
-		Version:  4,
-		TTL:      64,
-		Id:       1,
-		Protocol: layers.IPProtocolUDP,
-		SrcIP:    net.ParseIP(fragTestSrc),
-		DstIP:    net.ParseIP(fragTestDst),
-	}
-	udp := &layers.UDP{SrcPort: 40000, DstPort: layers.UDPPort(dstPort)}
-	require.NoError(tb, udp.SetNetworkLayerForChecksum(ip))
-
-	buf := gopacket.NewSerializeBuffer()
-	opts := gopacket.SerializeOptions{ComputeChecksums: true, FixLengths: true}
-	require.NoError(tb, gopacket.SerializeLayers(buf, opts, ip, udp, gopacket.Payload(make([]byte, payloadLen))))
-	return buf.Bytes()
-}
-
-func allowUDP(tb testing.TB, m *Manager, dstPort uint16) {
-	tb.Helper()
-	_, err := m.AddPeerFiltering(nil, net.ParseIP(fragTestSrc), fw.ProtocolUDP, nil,
-		&fw.Port{Values: []uint16{dstPort}}, fw.ActionAccept, "")
-	require.NoError(tb, err)
-}
-
-// TestFragment_TrailingWithoutFirstDropped is the core bypass repro: a trailing
-// fragment with no allowed first fragment on record must be dropped. Before the
-// fix, filterInbound returned false (allow) for any fragment.
-func TestFragment_TrailingWithoutFirstDropped(t *testing.T) {
-	m := newFragmentTestManager(t)
-
-	frag := trailingFragment(t, 0x1234, 185, false, 40)
-	require.True(t, m.filterInbound(frag, len(frag)),
-		"trailing fragment without an allowed first fragment must be dropped")
-}
-
-// TestFragment_AllowedFirstPassesTrailing verifies that once a first fragment
-// passes the ACL, its trailing fragments inherit the allow verdict.
-func TestFragment_AllowedFirstPassesTrailing(t *testing.T) {
-	m := newFragmentTestManager(t)
-	allowUDP(t, m, 8080)
-
-	// First fragment: UDP header (8) + 32 payload = 40 octets -> headerEnd = 5.
-	first := firstFragmentUDP(t, 0x2222, 8080, 32)
-	require.False(t, m.filterInbound(first, len(first)),
-		"allowed first fragment should pass and be recorded")
-
-	trailing := trailingFragment(t, 0x2222, 5, false, 24)
-	require.False(t, m.filterInbound(trailing, len(trailing)),
-		"trailing fragment of an allowed datagram should pass")
-}
-
-// TestFragment_DeniedFirstDropsTrailing verifies that a first fragment blocked
-// by the ACL leaves no verdict, so its trailing fragments are dropped.
-func TestFragment_DeniedFirstDropsTrailing(t *testing.T) {
-	m := newFragmentTestManager(t)
-	// No accept rule: local traffic defaults to deny.
-
-	first := firstFragmentUDP(t, 0x3333, 9999, 32)
-	require.True(t, m.filterInbound(first, len(first)),
-		"first fragment to a blocked port should be dropped by the ACL")
-
-	trailing := trailingFragment(t, 0x3333, 5, false, 24)
-	require.True(t, m.filterInbound(trailing, len(trailing)),
-		"trailing fragment of a denied datagram must be dropped")
-}
-
-// TestFragment_OverlappingHeaderDropped covers the RFC 1858 §4 / RFC 3128
-// overlapping-fragment rewrite: a trailing fragment starting inside the range
-// the ACL already inspected is dropped and poisons the datagram. TCP is used so
-// the overlap lands on real header bytes (the flags at byte 13).
-func TestFragment_OverlappingHeaderDropped(t *testing.T) {
-	m := newFragmentTestManager(t)
-	_, err := m.AddPeerFiltering(nil, net.ParseIP(fragTestSrc), fw.ProtocolTCP, nil,
-		&fw.Port{Values: []uint16{8080}}, fw.ActionAccept, "")
-	require.NoError(t, err)
-
-	// First fragment: TCP header (20) + 12 data = 32 bytes -> headerEnd = 4 octets.
-	first := firstFragmentTCP(t, 0x4444, 8080)
-	require.False(t, m.filterInbound(first, len(first)))
-
-	// Overlapping fragment at offset 1 (byte 8) falls inside the inspected TCP
-	// header, so it could rewrite the flags or port on reassembly.
-	overlap := trailingFragmentTo(t, fragTestDst, layers.IPProtocolTCP, 0x4444, 1, true, 32)
-	require.True(t, m.filterInbound(overlap, len(overlap)),
-		"fragment overlapping the inspected header must be dropped")
-
-	// The datagram is now poisoned: a later, non-overlapping fragment is also
-	// dropped because the verdict was removed.
-	later := trailingFragmentTo(t, fragTestDst, layers.IPProtocolTCP, 0x4444, 4, false, 24)
-	require.True(t, m.filterInbound(later, len(later)),
-		"fragments after an overlap must be dropped (datagram poisoned)")
-}
-
-// TestFragment_OffsetZeroOverlapPoisons covers the RFC 3128 offset-zero rewrite:
-// an allowed first fragment followed by a denied offset-zero fragment for the
-// same datagram must not leave the earlier allow verdict in place.
-func TestFragment_OffsetZeroOverlapPoisons(t *testing.T) {
-	m := newFragmentTestManager(t)
-	allowUDP(t, m, 8080)
-
-	allowed := firstFragmentUDP(t, 0x5A5A, 8080, 32)
-	require.False(t, m.filterInbound(allowed, len(allowed)),
-		"allowed first fragment should pass and be recorded")
-
-	// A second offset-zero fragment to a denied port supersedes the datagram's
-	// verdict; it is dropped and must not leave the allow in place.
-	denied := firstFragmentUDP(t, 0x5A5A, 9999, 32)
-	require.True(t, m.filterInbound(denied, len(denied)),
-		"denied offset-zero fragment must be dropped")
-
-	trailing := trailingFragment(t, 0x5A5A, 5, false, 24)
-	require.True(t, m.filterInbound(trailing, len(trailing)),
-		"trailing fragment must be denied after the datagram was poisoned")
-}
-
-// TestFragment_TinyFirstDropped covers the tiny-fragment attack: a first
-// fragment too small to contain the full transport header can't be
-// ACL-evaluated and must be dropped.
-func TestFragment_TinyFirstDropped(t *testing.T) {
-	m := newFragmentTestManager(t)
-	allowUDP(t, m, 8080)
-
-	// IPv4 header + 4 raw bytes, MF set, offset 0: too small for the 8-byte UDP
-	// header, so it decodes to L3 only.
-	tiny := trailingFragment(t, 0x5555, 0, true, 4)
-	require.True(t, m.filterInbound(tiny, len(tiny)),
-		"tiny first fragment without a full L4 header must be dropped")
-}
-
-// TestFragment_TCPFirstFragment verifies the TCP arm of the transport decode: a
-// first fragment carrying the full 20-byte TCP header is ACL-evaluated and its
-// trailing fragments inherit the verdict.
-func TestFragment_TCPFirstFragment(t *testing.T) {
-	m := newFragmentTestManager(t)
-	_, err := m.AddPeerFiltering(nil, net.ParseIP(fragTestSrc), fw.ProtocolTCP, nil,
-		&fw.Port{Values: []uint16{8080}}, fw.ActionAccept, "")
-	require.NoError(t, err)
-
-	// TCP header (20) + 12 data = 32 bytes -> headerEnd = 4 octets.
-	first := firstFragmentTCP(t, 0x6666, 8080)
-	require.False(t, m.filterInbound(first, len(first)),
-		"allowed TCP first fragment should pass and be recorded")
-
-	trailing := trailingFragmentTo(t, fragTestDst, layers.IPProtocolTCP, 0x6666, 4, false, 24)
-	require.False(t, m.filterInbound(trailing, len(trailing)),
-		"trailing fragment of an allowed TCP datagram should pass")
-}
-
-// TestFragment_TCPTinyFirstDropped verifies the TCP minimum header length: 12
-// bytes would satisfy a UDP header but falls short of the 20-byte TCP header.
-func TestFragment_TCPTinyFirstDropped(t *testing.T) {
-	m := newFragmentTestManager(t)
-	_, err := m.AddPeerFiltering(nil, net.ParseIP(fragTestSrc), fw.ProtocolTCP, nil,
-		&fw.Port{Values: []uint16{8080}}, fw.ActionAccept, "")
-	require.NoError(t, err)
-
-	tiny := trailingFragmentTo(t, fragTestDst, layers.IPProtocolTCP, 0x7777, 0, true, 12)
-	require.True(t, m.filterInbound(tiny, len(tiny)),
-		"first fragment shorter than the TCP header must be dropped")
-}
-
-// TestFragment_ConntrackAllowsFirstFragment verifies the conntrack branch: reply
-// fragments of an outbound-established UDP flow pass without any inbound rule.
-func TestFragment_ConntrackAllowsFirstFragment(t *testing.T) {
-	m := newFragmentTestManager(t)
-
-	out := outboundUDPPacket(t, 12345, 40000)
-	require.False(t, m.filterOutbound(out, len(out)))
-
-	first := firstFragmentUDP(t, 0x8888, 12345, 32)
-	require.False(t, m.filterInbound(first, len(first)),
-		"reply first fragment should pass via conntrack")
-
-	trailing := trailingFragment(t, 0x8888, 5, false, 24)
-	require.False(t, m.filterInbound(trailing, len(trailing)),
-		"trailing fragment of a tracked flow should pass")
-}
-
-// TestFragment_RoutingDisabledDropsFragment verifies routed first fragments are
-// dropped when routing is disabled.
-func TestFragment_RoutingDisabledDropsFragment(t *testing.T) {
-	m := newFragmentTestManager(t)
-	m.routingEnabled.Store(false)
-
-	first := firstFragmentUDPTo(t, "198.51.100.10", 0x9999, 8080, 32)
-	require.True(t, m.filterInbound(first, len(first)),
-		"routed first fragment must be dropped when routing is disabled")
-}
-
-// TestFragment_RouteACL verifies the route-ACL branch: fragments to a non-local
-// destination follow the route rules, allowed datagrams pass their trailing
-// fragments and denied ones don't.
-func TestFragment_RouteACL(t *testing.T) {
-	m := newFragmentTestManager(t)
-	m.routingEnabled.Store(true)
-	m.nativeRouter.Store(false)
-
-	_, err := m.AddRouteFiltering(
-		[]byte("rt-1"),
-		[]netip.Prefix{netip.MustParsePrefix("100.10.0.0/16")},
-		fw.Network{Prefix: netip.MustParsePrefix("198.51.100.0/24")},
-		fw.ProtocolUDP,
-		nil,
-		&fw.Port{Values: []uint16{8080}},
-		fw.ActionAccept,
-	)
-	require.NoError(t, err)
-
-	first := firstFragmentUDPTo(t, "198.51.100.10", 0xAAAA, 8080, 32)
-	require.False(t, m.filterInbound(first, len(first)),
-		"route-ACL-allowed first fragment should pass")
-	trailing := trailingFragmentTo(t, "198.51.100.10", layers.IPProtocolUDP, 0xAAAA, 5, false, 24)
-	require.False(t, m.filterInbound(trailing, len(trailing)),
-		"trailing fragment of an allowed routed datagram should pass")
-
-	denied := firstFragmentUDPTo(t, "198.51.100.10", 0xBBBB, 9999, 32)
-	require.True(t, m.filterInbound(denied, len(denied)),
-		"route-ACL-denied first fragment must be dropped")
-	deniedTrailing := trailingFragmentTo(t, "198.51.100.10", layers.IPProtocolUDP, 0xBBBB, 5, false, 24)
-	require.True(t, m.filterInbound(deniedTrailing, len(deniedTrailing)),
-		"trailing fragment of a denied routed datagram must be dropped")
-}
-
-// TestFragment_ExpiredVerdictDropsTrailing verifies a verdict older than the
-// tracker timeout no longer admits trailing fragments.
-func TestFragment_ExpiredVerdictDropsTrailing(t *testing.T) {
-	m := newFragmentTestManager(t)
-	allowUDP(t, m, 8080)
-
-	first := firstFragmentUDP(t, 0xCCCC, 8080, 32)
-	require.False(t, m.filterInbound(first, len(first)))
-
-	m.fragments.mutex.Lock()
-	for key, entry := range m.fragments.entries {
-		entry.recordedAt = time.Now().Add(-defaultFragmentTimeout - time.Second)
-		m.fragments.entries[key] = entry
-	}
-	m.fragments.mutex.Unlock()
-
-	trailing := trailingFragment(t, 0xCCCC, 5, false, 24)
-	require.True(t, m.filterInbound(trailing, len(trailing)),
-		"trailing fragment after verdict expiry must be dropped")
-}
-
-// TestFragment_CapacityFailsClosed verifies the table cap: at capacity, new
-// datagram verdicts are not recorded (their trailing fragments are dropped)
-// while already-recorded datagrams keep working.
-func TestFragment_CapacityFailsClosed(t *testing.T) {
-	m := newFragmentTestManager(t)
-	allowUDP(t, m, 8080)
-
-	m.fragments.mutex.Lock()
-	m.fragments.maxEntries = 1
-	m.fragments.mutex.Unlock()
-
-	first1 := firstFragmentUDP(t, 0x0101, 8080, 32)
-	require.False(t, m.filterInbound(first1, len(first1)))
-
-	first2 := firstFragmentUDP(t, 0x0202, 8080, 32)
-	require.False(t, m.filterInbound(first2, len(first2)),
-		"first fragment itself still passes at capacity")
-
-	trailing2 := trailingFragment(t, 0x0202, 5, false, 24)
-	require.True(t, m.filterInbound(trailing2, len(trailing2)),
-		"trailing fragment of an unrecorded datagram must be dropped at capacity")
-
-	trailing1 := trailingFragment(t, 0x0101, 5, false, 24)
-	require.False(t, m.filterInbound(trailing1, len(trailing1)),
-		"already-recorded datagram should keep passing at capacity")
-}
-
-// v6FragmentHeader builds the 8-byte IPv6 fragment extension header for the
-// given inner protocol, offset (8-byte units), More Fragments bit and id.
-func v6FragmentHeader(proto layers.IPProtocol, offsetOctets uint16, moreFragments bool, id uint32) []byte {
-	offsetFlags := offsetOctets << 3
-	if moreFragments {
-		offsetFlags |= 1
-	}
-	hdr := make([]byte, 8)
-	hdr[0] = uint8(proto)
-	binary.BigEndian.PutUint16(hdr[2:4], offsetFlags)
-	binary.BigEndian.PutUint32(hdr[4:8], id)
-	return hdr
-}
-
-func v6UDPHeader(dstPort uint16, dataLen int) []byte {
-	hdr := make([]byte, 8)
-	binary.BigEndian.PutUint16(hdr[0:2], 40000)
-	binary.BigEndian.PutUint16(hdr[2:4], dstPort)
-	binary.BigEndian.PutUint16(hdr[4:6], uint16(8+dataLen))
-	return hdr
-}
-
-// firstFragmentUDPv6 builds the first fragment of a fragmented IPv6 UDP
-// datagram: fragment header (offset 0, More Fragments set) + full UDP header +
-// data.
-func firstFragmentUDPv6(tb testing.TB, id uint32, dstPort uint16, dataLen int) []byte {
-	tb.Helper()
-	return fragmentUDPv6(tb, id, dstPort, dataLen, true)
-}
-
-// fragmentUDPv6 builds an offset-zero IPv6 UDP fragment. With moreFragments
-// false it is an atomic fragment (a complete datagram, RFC 6946).
-func fragmentUDPv6(tb testing.TB, id uint32, dstPort uint16, dataLen int, moreFragments bool) []byte {
-	tb.Helper()
-
-	ip := &layers.IPv6{
-		Version:    6,
-		NextHeader: layers.IPProtocolIPv6Fragment,
-		HopLimit:   64,
-		SrcIP:      net.ParseIP(fragTestSrcV6),
-		DstIP:      net.ParseIP(fragTestDstV6),
-	}
-	payload := append(v6FragmentHeader(layers.IPProtocolUDP, 0, moreFragments, id), v6UDPHeader(dstPort, dataLen)...)
-	payload = append(payload, make([]byte, dataLen)...)
-
-	buf := gopacket.NewSerializeBuffer()
-	require.NoError(tb, gopacket.SerializeLayers(buf, gopacket.SerializeOptions{FixLengths: true}, ip, gopacket.Payload(payload)))
-	return buf.Bytes()
-}
-
-// trailingFragmentV6 builds a non-first IPv6 fragment: fragment header at the
-// given offset carrying raw data and no transport header.
-func trailingFragmentV6(tb testing.TB, id uint32, offsetOctets uint16, moreFragments bool, dataLen int) []byte {
-	tb.Helper()
-
-	ip := &layers.IPv6{
-		Version:    6,
-		NextHeader: layers.IPProtocolIPv6Fragment,
-		HopLimit:   64,
-		SrcIP:      net.ParseIP(fragTestSrcV6),
-		DstIP:      net.ParseIP(fragTestDstV6),
-	}
-	payload := append(v6FragmentHeader(layers.IPProtocolUDP, offsetOctets, moreFragments, id), make([]byte, dataLen)...)
-
-	buf := gopacket.NewSerializeBuffer()
-	require.NoError(tb, gopacket.SerializeLayers(buf, gopacket.SerializeOptions{FixLengths: true}, ip, gopacket.Payload(payload)))
-	return buf.Bytes()
-}
-
-// TestFragmentV6_TrailingWithoutFirstDropped verifies the IPv6 bypass is closed:
-// a trailing fragment with no allowed first fragment is dropped.
-func TestFragmentV6_TrailingWithoutFirstDropped(t *testing.T) {
-	m := newFragmentTestManager(t)
-
-	frag := trailingFragmentV6(t, 0xAABBCCDD, 100, false, 40)
-	require.True(t, m.filterInbound(frag, len(frag)),
-		"IPv6 trailing fragment without an allowed first fragment must be dropped")
-}
-
-// TestFragmentV6_AllowedFirstPassesTrailing verifies IPv6 fragments are
-// evaluated like IPv4: an allowed first fragment lets its trailing fragments
-// through.
-func TestFragmentV6_AllowedFirstPassesTrailing(t *testing.T) {
-	m := newFragmentTestManager(t)
-	_, err := m.AddPeerFiltering(nil, net.ParseIP(fragTestSrcV6), fw.ProtocolUDP, nil,
-		&fw.Port{Values: []uint16{8080}}, fw.ActionAccept, "")
-	require.NoError(t, err)
-
-	// First fragment: UDP header (8) + 32 data = 40 octets -> headerEnd = 5.
-	first := firstFragmentUDPv6(t, 0xAABBCCDD, 8080, 32)
-	require.False(t, m.filterInbound(first, len(first)),
-		"allowed IPv6 first fragment should pass and be recorded")
-
-	trailing := trailingFragmentV6(t, 0xAABBCCDD, 5, false, 24)
-	require.False(t, m.filterInbound(trailing, len(trailing)),
-		"trailing fragment of an allowed IPv6 datagram should pass")
-}
-
-// TestFragmentV6_AtomicNotCached verifies an IPv6 atomic fragment (fragment
-// header with offset 0 and no More Fragments, a complete datagram per RFC 6946)
-// is evaluated but not recorded, so a flood of allowed atomic fragments can't
-// exhaust the verdict table.
-func TestFragmentV6_AtomicNotCached(t *testing.T) {
-	m := newFragmentTestManager(t)
-	_, err := m.AddPeerFiltering(nil, net.ParseIP(fragTestSrcV6), fw.ProtocolUDP, nil,
-		&fw.Port{Values: []uint16{8080}}, fw.ActionAccept, "")
-	require.NoError(t, err)
-
-	atomic := fragmentUDPv6(t, 0xA70301C, 8080, 16, false)
-	require.False(t, m.filterInbound(atomic, len(atomic)),
-		"allowed IPv6 atomic fragment should pass")
-
-	m.fragments.mutex.Lock()
-	n := len(m.fragments.entries)
-	m.fragments.mutex.Unlock()
-	require.Zero(t, n, "atomic fragment must not create a verdict entry")
-
-	// A genuine fragmented datagram (More Fragments set) is still recorded.
-	first := fragmentUDPv6(t, 0xBEEF, 8080, 32, true)
-	require.False(t, m.filterInbound(first, len(first)))
-	m.fragments.mutex.Lock()
-	n = len(m.fragments.entries)
-	m.fragments.mutex.Unlock()
-	require.Equal(t, 1, n, "genuine first fragment must record a verdict")
-}

client/iface/iface_test.go

@@ -464,8 +464,6 @@ func Test_RemovePeer(t *testing.T) {
 }
 
 func Test_ConnectPeers(t *testing.T) {
-	t.Setenv("NB_DISABLE_EBPF_WG_PROXY", "true")
-
 	peer1ifaceName := fmt.Sprintf("utun%d", WgIntNumber+400)
 	peer1wgIP := netip.MustParsePrefix("10.99.99.17/30")
 	peer1Key, _ := wgtypes.GeneratePrivateKey()
@@ -507,8 +505,12 @@ func Test_ConnectPeers(t *testing.T) {
 		t.Fatal(err)
 	}
 
-	localIP1 := "127.0.0.1"
+	localIP, err := getLocalIP()
-	peer1endpoint, err := net.ResolveUDPAddr("udp", fmt.Sprintf("%s:%d", localIP1, peer1wgPort))
+	if err != nil {
+		t.Fatal(err)
+	}
+
+	peer1endpoint, err := net.ResolveUDPAddr("udp", fmt.Sprintf("%s:%d", localIP, peer1wgPort))
 	if err != nil {
 		t.Fatal(err)
 	}
@@ -544,8 +546,7 @@ func Test_ConnectPeers(t *testing.T) {
 		t.Fatal(err)
 	}
 
-	localIP2 := "127.0.0.1"
+	peer2endpoint, err := net.ResolveUDPAddr("udp", fmt.Sprintf("%s:%d", localIP, peer2wgPort))
-	peer2endpoint, err := net.ResolveUDPAddr("udp", fmt.Sprintf("%s:%d", localIP2, peer2wgPort))
 	if err != nil {
 		t.Fatal(err)
 	}
@@ -568,17 +569,17 @@ func Test_ConnectPeers(t *testing.T) {
 	if err != nil {
 		t.Fatal(err)
 	}
-	// The peers use userspace WireGuard (stdnet transport). A tight busy-loop
+	// todo: investigate why in some tests execution we need 30s
-	// here starves the wireguard-go goroutines that process the handshake, so
-	// poll on a ticker instead and yield the CPU between checks. WireGuard also
-	// only retries a lost handshake initiation every REKEY_TIMEOUT (5s), which
-	// is why the overall wait can occasionally stretch to tens of seconds.
 	timeout := 30 * time.Second
 	timeoutChannel := time.After(timeout)
-	ticker := time.NewTicker(500 * time.Millisecond)
-	defer ticker.Stop()
 
 	for {
+		select {
+		case <-timeoutChannel:
+			t.Fatalf("waiting for peer handshake timeout after %s", timeout.String())
+		default:
+		}
+
 		peer, gpErr := getPeer(peer1ifaceName, peer2Key.PublicKey().String())
 		if gpErr != nil {
 			t.Fatal(gpErr)
@@ -587,12 +588,6 @@ func Test_ConnectPeers(t *testing.T) {
 			t.Log("peers successfully handshake")
 			break
 		}
-
-		select {
-		case <-timeoutChannel:
-			t.Fatalf("waiting for peer handshake timeout after %s", timeout.String())
-		case <-ticker.C:
-		}
 	}
 
 }
@@ -620,3 +615,28 @@ func getPeer(ifaceName, peerPubKey string) (wgtypes.Peer, error) {
 	}
 	return wgtypes.Peer{}, fmt.Errorf("peer not found")
 }
+
+func getLocalIP() (string, error) {
+	// Get all interfaces
+	addrs, err := net.InterfaceAddrs()
+	if err != nil {
+		return "", err
+	}
+
+	for _, addr := range addrs {
+		ipNet, ok := addr.(*net.IPNet)
+		if !ok {
+			continue
+		}
+		if ipNet.IP.IsLoopback() {
+			continue
+		}
+
+		if ipNet.IP.To4() == nil {
+			continue
+		}
+		return ipNet.IP.String(), nil
+	}
+
+	return "", fmt.Errorf("no local IP found")
+}

client/iface/netstack/env.go

@@ -3,31 +3,14 @@
 package netstack
 
 import (
-	"net"
+	"fmt"
 	"os"
 	"strconv"
 
 	log "github.com/sirupsen/logrus"
 )
 
-const (
+const EnvUseNetstackMode = "NB_USE_NETSTACK_MODE"
-	EnvUseNetstackMode = "NB_USE_NETSTACK_MODE"
-
-	// EnvSocks5ListenerPort overrides the port the SOCKS5 proxy listens on.
-	EnvSocks5ListenerPort = "NB_SOCKS5_LISTENER_PORT"
-
-	// EnvSocks5ListenerAddress overrides the host/IP the SOCKS5 proxy binds to.
-	// The proxy is a bridge for local host applications into the userspace
-	// WireGuard netstack, so it binds to loopback by default. Override this only
-	// when the proxy must be reachable from other hosts (e.g. a container
-	// gateway); doing so exposes an unauthenticated SOCKS5 proxy on that
-	// address.
-	EnvSocks5ListenerAddress = "NB_SOCKS5_LISTENER_ADDRESS"
-
-	// defaultSocks5Host is the loopback address the SOCKS5 proxy binds to unless
-	// overridden via EnvSocks5ListenerAddress.
-	defaultSocks5Host = "127.0.0.1"
-)
 
 // IsEnabled todo: move these function to cmd layer
 func IsEnabled() bool {
@@ -35,40 +18,24 @@ func IsEnabled() bool {
 }
 
 func ListenAddr() string {
-	return net.JoinHostPort(listenHost(), strconv.Itoa(listenPort()))
+	sPort := os.Getenv("NB_SOCKS5_LISTENER_PORT")
-}
-
-// listenHost returns the host/IP the SOCKS5 proxy binds to. It defaults to
-// loopback and only honors EnvSocks5ListenerAddress when it holds a valid IP.
-func listenHost() string {
-	addr := os.Getenv(EnvSocks5ListenerAddress)
-	if addr == "" {
-		return defaultSocks5Host
-	}
-	if net.ParseIP(addr) == nil {
-		log.Warnf("invalid socks5 listener address %q, falling back to default: %s", addr, defaultSocks5Host)
-		return defaultSocks5Host
-	}
-	return addr
-}
-
-// listenPort returns the port the SOCKS5 proxy binds to, defaulting to
-// DefaultSocks5Port when EnvSocks5ListenerPort is unset or invalid.
-func listenPort() int {
-	sPort := os.Getenv(EnvSocks5ListenerPort)
 	if sPort == "" {
-		return DefaultSocks5Port
+		return listenAddr(DefaultSocks5Port)
 	}
 
 	port, err := strconv.Atoi(sPort)
 	if err != nil {
 		log.Warnf("invalid socks5 listener port, unable to convert it to int, falling back to default: %d", DefaultSocks5Port)
-		return DefaultSocks5Port
+		return listenAddr(DefaultSocks5Port)
 	}
 	if port < 1 || port > 65535 {
 		log.Warnf("invalid socks5 listener port, it should be in the range 1-65535, falling back to default: %d", DefaultSocks5Port)
-		return DefaultSocks5Port
+		return listenAddr(DefaultSocks5Port)
 	}
 
-	return port
+	return listenAddr(port)
+}
+
+func listenAddr(port int) string {
+	return fmt.Sprintf("0.0.0.0:%d", port)
 }

client/iface/netstack/env_test.go

@@ -1,63 +0,0 @@
-//go:build !js
-
-package netstack
-
-import (
-	"net"
-	"strconv"
-	"testing"
-)
-
-func TestListenAddr_DefaultsToLoopback(t *testing.T) {
-	// No env overrides: must bind loopback, never all interfaces.
-	got := ListenAddr()
-	want := net.JoinHostPort("127.0.0.1", strconv.Itoa(DefaultSocks5Port))
-	if got != want {
-		t.Fatalf("ListenAddr() = %q, want %q", got, want)
-	}
-}
-
-func TestListenAddr_AddressOverride(t *testing.T) {
-	tests := []struct {
-		name string
-		env  string
-		want string
-	}{
-		{name: "valid override honored", env: "0.0.0.0", want: "0.0.0.0"},
-		{name: "valid specific ip honored", env: "10.0.0.5", want: "10.0.0.5"},
-		{name: "ipv6 loopback bracketed", env: "::1", want: "::1"},
-		{name: "invalid falls back to loopback", env: "not-an-ip", want: "127.0.0.1"},
-		{name: "empty falls back to loopback", env: "", want: "127.0.0.1"},
-	}
-	for _, tc := range tests {
-		t.Run(tc.name, func(t *testing.T) {
-			t.Setenv(EnvSocks5ListenerAddress, tc.env)
-			want := net.JoinHostPort(tc.want, strconv.Itoa(DefaultSocks5Port))
-			if got := ListenAddr(); got != want {
-				t.Fatalf("ListenAddr() = %q, want %q", got, want)
-			}
-		})
-	}
-}
-
-func TestListenAddr_PortOverride(t *testing.T) {
-	tests := []struct {
-		name string
-		env  string
-		want int
-	}{
-		{name: "valid port honored", env: "1081", want: 1081},
-		{name: "non-numeric falls back", env: "abc", want: DefaultSocks5Port},
-		{name: "out of range falls back", env: "70000", want: DefaultSocks5Port},
-		{name: "zero falls back", env: "0", want: DefaultSocks5Port},
-	}
-	for _, tc := range tests {
-		t.Run(tc.name, func(t *testing.T) {
-			t.Setenv(EnvSocks5ListenerPort, tc.env)
-			want := net.JoinHostPort("127.0.0.1", strconv.Itoa(tc.want))
-			if got := ListenAddr(); got != want {
-				t.Fatalf("ListenAddr() = %q, want %q", got, want)
-			}
-		})
-	}
-}

client/installer.nsis

@@ -6,7 +6,7 @@
 !define DESCRIPTION "Connect your devices into a secure WireGuard-based overlay network with SSO, MFA, and granular access controls."
 !define INSTALLER_NAME "netbird-installer.exe"
 !define MAIN_APP_EXE "Netbird"
-!define ICON "ui\\build\\windows\\icon.ico"
+!define ICON "ui\\assets\\netbird.ico"
 !define BANNER "ui\\build\\banner.bmp"
 !define LICENSE_DATA "..\\LICENSE"
 
@@ -79,6 +79,8 @@ ShowInstDetails Show
 
 !insertmacro MUI_PAGE_DIRECTORY
 
+Page custom AutostartPage AutostartPageLeave
+
 !insertmacro MUI_PAGE_INSTFILES
 
 !insertmacro MUI_PAGE_FINISH
@@ -95,12 +97,40 @@ UninstPage custom un.DeleteDataPage un.DeleteDataPageLeave
 
 !insertmacro MUI_LANGUAGE "English"
 
+; Variables for autostart option
+Var AutostartCheckbox
+Var AutostartEnabled
+
 ; Variables for uninstall data deletion option
 Var DeleteDataCheckbox
 Var DeleteDataEnabled
 
 ######################################################################
 
+; Function to create the autostart options page
+Function AutostartPage
+  !insertmacro MUI_HEADER_TEXT "Startup Options" "Configure how ${APP_NAME} launches with Windows."
+
+  nsDialogs::Create 1018
+  Pop $0
+
+  ${If} $0 == error
+    Abort
+  ${EndIf}
+
+  ${NSD_CreateCheckbox} 0 20u 100% 10u "Start ${APP_NAME} UI automatically when Windows starts"
+  Pop $AutostartCheckbox
+  ${NSD_Check} $AutostartCheckbox
+  StrCpy $AutostartEnabled "1"
+
+  nsDialogs::Show
+FunctionEnd
+
+; Function to handle leaving the autostart page
+Function AutostartPageLeave
+  ${NSD_GetState} $AutostartCheckbox $AutostartEnabled
+FunctionEnd
+
 ; Function to create the uninstall data deletion page
 Function un.DeleteDataPage
   !insertmacro MUI_HEADER_TEXT "Uninstall Options" "Choose whether to delete ${APP_NAME} data."
@@ -171,6 +201,8 @@ Pop $0
 
 Function .onInit
 StrCpy $INSTDIR "${INSTALL_DIR}"
+; Default autostart to enabled so silent installs (/S) match the interactive default
+StrCpy $AutostartEnabled "1"
 
 ; Pre-0.70.1 installers ran without SetRegView, so their uninstall keys live
 ; in the 32-bit view. Fall back to it so upgrades still find them.
@@ -228,12 +260,17 @@ WriteRegStr ${REG_ROOT} "${UNINSTALL_PATH}"  "Publisher" "${COMP_NAME}"
 
 WriteRegStr ${REG_ROOT} "${UI_REG_APP_PATH}" "" "$INSTDIR\${UI_APP_EXE}"
 
-; Autostart is owned by the UI's per-user setting (HKCU\...\Run via Wails),
+; Create autostart registry entry based on checkbox
-; not the installer. Drop the machine-wide entry older installers wrote so the
+DetailPrint "Autostart enabled: $AutostartEnabled"
-; toggle is the single source of truth. HKCU is left untouched -- it may hold
+${If} $AutostartEnabled == "1"
-; the user's own toggle state, which must survive upgrades.
+    WriteRegStr HKLM "${AUTOSTART_REG_KEY}" "${APP_NAME}" '"$INSTDIR\${UI_APP_EXE}.exe"'
-DetailPrint "Removing installer-managed autostart registry entry if present..."
+    DetailPrint "Added autostart registry entry: $INSTDIR\${UI_APP_EXE}.exe"
-DeleteRegValue HKLM "${AUTOSTART_REG_KEY}" "${APP_NAME}"
+${Else}
+    DeleteRegValue HKLM "${AUTOSTART_REG_KEY}" "${APP_NAME}"
+    ; Legacy: pre-HKLM installs wrote to HKCU; clean that up too.
+    DeleteRegValue HKCU "${AUTOSTART_REG_KEY}" "${APP_NAME}"
+    DetailPrint "Autostart not enabled by user"
+${EndIf}
 
 EnVar::SetHKLM
 EnVar::AddValueEx "path" "$INSTDIR"
@@ -243,43 +280,6 @@ CreateShortCut "$SMPROGRAMS\${APP_NAME}.lnk" "$INSTDIR\${UI_APP_EXE}"
 CreateShortCut "$DESKTOP\${APP_NAME}.lnk" "$INSTDIR\${UI_APP_EXE}"
 SectionEnd
 
-# Install the Microsoft Edge WebView2 runtime if it isn't already present.
-# Macro adapted from Wails3's NSIS template (wails_tools.nsh): a registry
-# probe followed by a silent install of the embedded evergreen bootstrapper.
-# The MicrosoftEdgeWebview2Setup.exe payload is staged next to this script
-# by the sign-pipelines build step (`wails3 generate webview2bootstrapper`).
-!macro nb.webview2runtime
-    SetRegView 64
-    # Per-machine install marker — populated when the runtime ships with
-    # Edge or has been installed by an admin previously.
-    ReadRegStr $0 HKLM "SOFTWARE\WOW6432Node\Microsoft\EdgeUpdate\Clients\{F3017226-FE2A-4295-8BDF-00C3A9A7E4C5}" "pv"
-    ${If} $0 != ""
-        Goto webview2_ok
-    ${EndIf}
-    # Per-user fallback for HKCU installs.
-    ReadRegStr $0 HKCU "Software\Microsoft\EdgeUpdate\Clients\{F3017226-FE2A-4295-8BDF-00C3A9A7E4C5}" "pv"
-    ${If} $0 != ""
-        Goto webview2_ok
-    ${EndIf}
-
-    SetDetailsPrint both
-    DetailPrint "Installing: WebView2 Runtime"
-    SetDetailsPrint listonly
-
-    InitPluginsDir
-    CreateDirectory "$pluginsdir\webview2bootstrapper"
-    SetOutPath "$pluginsdir\webview2bootstrapper"
-    File "MicrosoftEdgeWebview2Setup.exe"
-    ExecWait '"$pluginsdir\webview2bootstrapper\MicrosoftEdgeWebview2Setup.exe" /silent /install'
-
-    SetDetailsPrint both
-    webview2_ok:
-!macroend
-
-Section -WebView2
-    !insertmacro nb.webview2runtime
-SectionEnd
-
 Section -Post
 ExecWait '"$INSTDIR\${MAIN_APP_EXE}" service install'
 ExecWait '"$INSTDIR\${MAIN_APP_EXE}" service start'
@@ -299,14 +299,11 @@ ExecWait '"$INSTDIR\${MAIN_APP_EXE}" service uninstall'
 DetailPrint "Terminating Netbird UI process..."
 ExecWait `taskkill /im ${UI_APP_EXE}.exe /f`
 
-; Remove autostart registry entries
+; Remove autostart registry entry
-DetailPrint "Removing autostart registry entries if they exist..."
+DetailPrint "Removing autostart registry entry if exists..."
-; Legacy machine-wide entry written by older installers.
 DeleteRegValue HKLM "${AUTOSTART_REG_KEY}" "${APP_NAME}"
-; Per-user entry the UI toggle writes via Wails (value name is the lowercase
+; Legacy: pre-HKLM installs wrote to HKCU; clean that up too.
-; app-name slug). Uninstall removes the app, so drop it too.
 DeleteRegValue HKCU "${AUTOSTART_REG_KEY}" "${APP_NAME}"
-DeleteRegValue HKCU "${AUTOSTART_REG_KEY}" "netbird"
 
 ; Handle data deletion based on checkbox
 DetailPrint "Checking if user requested data deletion..."
@@ -329,9 +326,9 @@ DetailPrint "Deleting application files..."
 Delete "$INSTDIR\${UI_APP_EXE}"
 Delete "$INSTDIR\${MAIN_APP_EXE}"
 Delete "$INSTDIR\wintun.dll"
-# Legacy: pre-Wails installs shipped opengl32.dll (Mesa3D for Fyne); remove
+!if ${ARCH} == "amd64"
-# any leftover copy on uninstall so old upgrades don't leave it behind.
 Delete "$INSTDIR\opengl32.dll"
+!endif
 DetailPrint "Removing application directory..."
 RmDir /r "$INSTDIR"
 

client/internal/auth/auth.go

@@ -3,7 +3,6 @@ package auth
 import (
 	"context"
 	"net/url"
-	"strings"
 	"sync"
 	"time"
 
@@ -22,25 +21,6 @@ import (
 	mgmProto "github.com/netbirdio/netbird/shared/management/proto"
 )
 
-// peerLoginExpiredMsg is the exact phrase the management server returns
-// when a previously SSO-enrolled peer's login has expired. Sourced from
-// shared/management/status/error.go (NewPeerLoginExpiredError). Matched
-// by substring so a future server-side rewording that keeps the phrase
-// still triggers the friendly fallback in Login().
-const peerLoginExpiredMsg = "peer login has expired"
-
-// errSetupKeyOnSSOExpiredPeer replaces the raw management error when the
-// user runs `netbird login -k <setup-key>` against a peer that was
-// originally enrolled via SSO. Wrapped in a PermissionDenied gRPC status
-// so callers' existing isPermissionDenied / isAuthError checks still
-// classify it correctly (early-exit from retry backoff, StatusNeedsLogin
-// in the server state machine).
-var errSetupKeyOnSSOExpiredPeer = status.Error(
-	codes.PermissionDenied,
-	"this peer was originally enrolled via SSO and its session has expired. "+
-		"Setup keys can only enrol new peers — run `netbird up` (interactive SSO) to re-login.",
-)
-
 // Auth manages authentication operations with the management server
 // It maintains a long-lived connection and automatically handles reconnection with backoff
 type Auth struct {
@@ -204,15 +184,6 @@ func (a *Auth) Login(ctx context.Context, setupKey string, jwtToken string) (err
 			log.Debugf("peer registration required")
 			_, err = a.registerPeer(client, ctx, setupKey, jwtToken, pubSSHKey)
 			if err != nil {
-				// The peer pub-key is already on file with the management
-				// server (originally enrolled via SSO) and the session has
-				// expired. The setup-key path can only enrol new peers, so
-				// retrying with -k will keep failing. Replace the raw mgm
-				// message with an actionable hint that tells the user to
-				// re-authenticate via SSO instead.
-				if setupKey != "" && jwtToken == "" && isPeerLoginExpired(err) {
-					err = errSetupKeyOnSSOExpiredPeer
-				}
 				isAuthError = isPermissionDenied(err)
 				return err
 			}
@@ -351,7 +322,6 @@ func (a *Auth) setSystemInfoFlags(info *system.Info) {
 		a.config.BlockLANAccess,
 		a.config.BlockInbound,
 		a.config.DisableIPv6,
-		a.config.SyncMessageVersion,
 		a.config.EnableSSHRoot,
 		a.config.EnableSSHSFTP,
 		a.config.EnableSSHLocalPortForwarding,
@@ -503,16 +473,3 @@ func isLoginNeeded(err error) bool {
 func isRegistrationNeeded(err error) bool {
 	return isPermissionDenied(err)
 }
-
-// isPeerLoginExpired reports whether err is the management server's
-// "peer login has expired" PermissionDenied response. Used by Login to
-// detect the case where the caller passed a setup-key but the peer is
-// actually an SSO-enrolled record whose session needs refreshing — the
-// setup-key path cannot help there.
-func isPeerLoginExpired(err error) bool {
-	if !isPermissionDenied(err) {
-		return false
-	}
-	s, _ := status.FromError(err)
-	return strings.Contains(s.Message(), peerLoginExpiredMsg)
-}

client/internal/auth/auth_test.go

@@ -1,80 +0,0 @@
-package auth
-
-import (
-	"errors"
-	"strings"
-	"testing"
-
-	"google.golang.org/grpc/codes"
-	"google.golang.org/grpc/status"
-)
-
-func TestIsPeerLoginExpired(t *testing.T) {
-	cases := []struct {
-		name string
-		err  error
-		want bool
-	}{
-		{
-			name: "nil",
-			err:  nil,
-			want: false,
-		},
-		{
-			name: "plain error (not a gRPC status)",
-			err:  errors.New("network read: connection reset"),
-			want: false,
-		},
-		{
-			name: "PermissionDenied with different message",
-			err:  status.Error(codes.PermissionDenied, "user is blocked"),
-			want: false,
-		},
-		{
-			name: "Unauthenticated with the expected phrase",
-			// Wrong status code — must still return false.
-			err:  status.Error(codes.Unauthenticated, "peer login has expired, please log in once more"),
-			want: false,
-		},
-		{
-			name: "exact server message",
-			err:  status.Error(codes.PermissionDenied, "peer login has expired, please log in once more"),
-			want: true,
-		},
-		{
-			name: "phrase as substring",
-			// Future-proofing: if mgm reworords but keeps the phrase,
-			// the friendly fallback must still kick in.
-			err:  status.Error(codes.PermissionDenied, "session refused: peer login has expired (account=foo)"),
-			want: true,
-		},
-	}
-
-	for _, tc := range cases {
-		t.Run(tc.name, func(t *testing.T) {
-			if got := isPeerLoginExpired(tc.err); got != tc.want {
-				t.Fatalf("isPeerLoginExpired(%v) = %v, want %v", tc.err, got, tc.want)
-			}
-		})
-	}
-}
-
-func TestErrSetupKeyOnSSOExpiredPeer(t *testing.T) {
-	// Sentinel must surface as PermissionDenied so the upstream
-	// isPermissionDenied / isAuthError checks classify it correctly
-	// (short-circuit retry backoff, set StatusNeedsLogin).
-	if !isPermissionDenied(errSetupKeyOnSSOExpiredPeer) {
-		t.Fatalf("errSetupKeyOnSSOExpiredPeer must be a PermissionDenied gRPC error")
-	}
-
-	// Message must actually mention SSO and `netbird up` so it is
-	// actionable for the end user. Loose substring checks keep the
-	// test resilient to copy edits.
-	s, _ := status.FromError(errSetupKeyOnSSOExpiredPeer)
-	msg := strings.ToLower(s.Message())
-	for _, want := range []string{"sso", "netbird up"} {
-		if !strings.Contains(msg, want) {
-			t.Errorf("sentinel message should contain %q, got %q", want, s.Message())
-		}
-	}
-}

client/internal/auth/device_flow.go

@@ -259,18 +259,12 @@ func (d *DeviceAuthorizationFlow) WaitToken(ctx context.Context, info AuthFlowIn
 	ticker := time.NewTicker(interval)
 	defer ticker.Stop()
 
-	log.Infof("device flow: waiting for user authorization, polling token endpoint every %s, code expires in %s", interval, timeout)
-
-	start := time.Now()
-	polls := 0
-
 	for {
 		select {
 		case <-waitCtx.Done():
 			return TokenInfo{}, waitCtx.Err()
 		case <-ticker.C:
 
-			polls++
 			tokenResponse, err := d.requestToken(info)
 			if err != nil {
 				return TokenInfo{}, fmt.Errorf("parsing token response failed with error: %v", err)
@@ -278,12 +272,10 @@ func (d *DeviceAuthorizationFlow) WaitToken(ctx context.Context, info AuthFlowIn
 
 			if tokenResponse.Error != "" {
 				if tokenResponse.Error == "authorization_pending" {
-					log.Tracef("device flow: authorization still pending after poll %d", polls)
 					continue
 				} else if tokenResponse.Error == "slow_down" {
 					interval += (3 * time.Second)
 					ticker.Reset(interval)
-					log.Infof("device flow: IdP requested slow_down, polling interval increased to %s", interval)
 					continue
 				}
 
@@ -299,12 +291,11 @@ func (d *DeviceAuthorizationFlow) WaitToken(ctx context.Context, info AuthFlowIn
 				UseIDToken:   d.providerConfig.UseIDToken,
 			}
 
-			err = validateTokenAudience(tokenInfo.GetTokenToUse(), d.providerConfig.Audience)
+			err = isValidAccessToken(tokenInfo.GetTokenToUse(), d.providerConfig.Audience)
 			if err != nil {
 				return TokenInfo{}, fmt.Errorf("validate access token failed with error: %v", err)
 			}
 
-			log.Infof("device flow: user authorization confirmed after %d polls in %s", polls, time.Since(start).Round(time.Second))
 			return tokenInfo, err
 		}
 	}

client/internal/auth/pending_flow.go

@@ -1,89 +0,0 @@
-package auth
-
-import (
-	"context"
-	"sync"
-	"time"
-)
-
-// PendingFlow stores an in-progress OAuth flow between the RPC that
-// initiates it (returns the verification URI to the UI) and the RPC
-// that waits for the user to complete it. The flow handle, the
-// device-code info, and the absolute expiry are kept together so the
-// waiting RPC can validate the device code and reuse the same flow.
-//
-// PendingFlow is safe for concurrent use; callers must not access the
-// stored fields directly.
-type PendingFlow struct {
-	mu         sync.Mutex
-	flow       OAuthFlow
-	info       AuthFlowInfo
-	expiresAt  time.Time
-	waitCancel context.CancelFunc
-}
-
-// NewPendingFlow returns an empty PendingFlow ready to be populated by Set.
-func NewPendingFlow() *PendingFlow {
-	return &PendingFlow{}
-}
-
-// Set stores the flow and its authorization info, computing the absolute
-// expiry from info.ExpiresIn (seconds, as returned by the IdP).
-func (p *PendingFlow) Set(flow OAuthFlow, info AuthFlowInfo) {
-	p.mu.Lock()
-	defer p.mu.Unlock()
-	p.flow = flow
-	p.info = info
-	p.expiresAt = time.Now().Add(time.Duration(info.ExpiresIn) * time.Second)
-}
-
-// Get returns the stored flow, info, and whether a flow is currently
-// pending. Returns (nil, zero, false) after Clear or before Set.
-func (p *PendingFlow) Get() (OAuthFlow, AuthFlowInfo, bool) {
-	p.mu.Lock()
-	defer p.mu.Unlock()
-	if p.flow == nil {
-		return nil, AuthFlowInfo{}, false
-	}
-	return p.flow, p.info, true
-}
-
-// ExpiresAt returns the absolute expiry of the pending flow. Returns
-// the zero time when no flow is pending.
-func (p *PendingFlow) ExpiresAt() time.Time {
-	p.mu.Lock()
-	defer p.mu.Unlock()
-	return p.expiresAt
-}
-
-// SetWaitCancel records the cancel function for the goroutine currently
-// blocked in WaitToken so a new RequestAuth can preempt it.
-func (p *PendingFlow) SetWaitCancel(cancel context.CancelFunc) {
-	p.mu.Lock()
-	defer p.mu.Unlock()
-	p.waitCancel = cancel
-}
-
-// CancelWait invokes and clears the stored wait-cancel, if any. Safe to
-// call when no wait is in progress.
-func (p *PendingFlow) CancelWait() {
-	p.mu.Lock()
-	cancel := p.waitCancel
-	p.waitCancel = nil
-	p.mu.Unlock()
-	if cancel != nil {
-		cancel()
-	}
-}
-
-// Clear resets the pending flow to empty. Any stored wait-cancel is
-// dropped without being invoked — call CancelWait first if the waiting
-// goroutine must be stopped.
-func (p *PendingFlow) Clear() {
-	p.mu.Lock()
-	defer p.mu.Unlock()
-	p.flow = nil
-	p.info = AuthFlowInfo{}
-	p.expiresAt = time.Time{}
-	p.waitCancel = nil
-}

client/internal/auth/pkce_flow.go

@@ -188,8 +188,6 @@ func (p *PKCEAuthorizationFlow) WaitToken(ctx context.Context, info AuthFlowInfo
 	waitCtx, cancel := context.WithTimeout(ctx, timeout)
 	defer cancel()
 
-	log.Infof("pkce flow: waiting for authorization callback on %s, timeout %s", p.oAuthConfig.RedirectURL, timeout)
-
 	tokenChan := make(chan *oauth2.Token, 1)
 	errChan := make(chan error, 1)
 
@@ -223,7 +221,6 @@ func (p *PKCEAuthorizationFlow) WaitToken(ctx context.Context, info AuthFlowInfo
 func (p *PKCEAuthorizationFlow) startServer(server *http.Server, tokenChan chan<- *oauth2.Token, errChan chan<- error) {
 	mux := http.NewServeMux()
 	mux.HandleFunc("/", func(w http.ResponseWriter, req *http.Request) {
-		log.Infof("pkce flow: received authorization callback from IdP")
 		cert := p.providerConfig.ClientCertPair
 		if cert != nil {
 			tr := &http.Transport{
@@ -274,18 +271,11 @@ func (p *PKCEAuthorizationFlow) handleRequest(req *http.Request) (*oauth2.Token,
 		return nil, fmt.Errorf("authentication failed: missing code")
 	}
 
-	exchangeStart := time.Now()
+	return p.oAuthConfig.Exchange(
-	token, err := p.oAuthConfig.Exchange(
 		req.Context(),
 		code,
 		oauth2.SetAuthURLParam("code_verifier", p.codeVerifier),
 	)
-	if err != nil {
-		return nil, err
-	}
-
-	log.Infof("pkce flow: authorization code exchanged for token in %s", time.Since(exchangeStart).Round(time.Millisecond))
-	return token, nil
 }
 
 func (p *PKCEAuthorizationFlow) parseOAuthToken(token *oauth2.Token) (TokenInfo, error) {
@@ -306,7 +296,7 @@ func (p *PKCEAuthorizationFlow) parseOAuthToken(token *oauth2.Token) (TokenInfo,
 		audience = p.providerConfig.ClientID
 	}
 
-	if err := validateTokenAudience(tokenInfo.GetTokenToUse(), audience); err != nil {
+	if err := isValidAccessToken(tokenInfo.GetTokenToUse(), audience); err != nil {
 		return TokenInfo{}, fmt.Errorf("authentication failed: invalid access token - %w", err)
 	}
 
@@ -320,11 +310,6 @@ func (p *PKCEAuthorizationFlow) parseOAuthToken(token *oauth2.Token) (TokenInfo,
 	return tokenInfo, nil
 }
 
-// parseEmailFromIDToken extracts the email (or name) claim from an ID token
-// without verifying its signature. The value is best-effort and used only as a
-// UX convenience (login hint prefill and display); it never drives an
-// authorization decision. The authoritative identity is established server-side
-// from the signature-verified token.
 func parseEmailFromIDToken(token string) (string, error) {
 	parts := strings.Split(token, ".")
 	if len(parts) < 2 {

client/internal/auth/sessionwatch/event.go

@@ -1,82 +0,0 @@
-package sessionwatch
-
-import (
-	"strconv"
-	"time"
-)
-
-// internal event kinds are no longer exposed: the watcher drives the Sink
-// directly (NotifyStateChange on deadline change/clear, PublishEvent at
-// each warning lead). Tests use a mock Sink to observe what the watcher
-// emits.
-
-// Metadata keys attached by the daemon to session-warning SystemEvents.
-// The UI tray reads these to build a locale-aware notification without
-// relying on the daemon's locale-less UserMessage string, and to
-// disambiguate the T-WarningLead notification from the T-FinalWarningLead
-// fallback that auto-opens the SessionAboutToExpire dialog.
-const (
-	// MetaSessionWarning is set to "true" on both warning events (T-10 and
-	// T-2) so the UI can detect a session-warning SystemEvent without
-	// matching on the message text. Use MetaSessionFinal to distinguish
-	// the two.
-	MetaSessionWarning = "session_warning"
-	// MetaSessionFinal is set to "true" on the T-FinalWarningLead event
-	// only. Consumers that need to auto-open the SessionAboutToExpire
-	// dialog gate on this; T-WarningLead events leave the field unset.
-	MetaSessionFinal = "session_final_warning"
-	// MetaSessionExpiresAt carries the absolute UTC deadline encoded with
-	// FormatExpiresAt; consumers must decode with ParseExpiresAt so a
-	// future format change stays a single edit.
-	MetaSessionExpiresAt = "session_expires_at"
-	// MetaSessionLeadMinutes carries the lead in whole minutes (WarningLead
-	// for the T-10 event, FinalWarningLead for the T-2 event) so the UI
-	// can show "expires in ~N minutes" without hardcoding either constant.
-	MetaSessionLeadMinutes = "lead_minutes"
-	// MetaSessionDeadlineRejected is attached to the ERROR/AUTHENTICATION
-	// SystemEvent the daemon emits when it discards a deadline from the
-	// management server (pre-epoch, too far in the future, or past the
-	// clock-skew tolerance). The value is the rejection reason string.
-	// userMessage is left empty; the UI detects the event via this key
-	// and builds a localized notification — same pattern as the session
-	// warnings above.
-	MetaSessionDeadlineRejected = "session_deadline_rejected"
-)
-
-// expiresAtLayout is the wire format used for MetaSessionExpiresAt.
-// Producer and consumers both go through FormatExpiresAt/ParseExpiresAt
-// so this layout stays a single source of truth.
-const expiresAtLayout = time.RFC3339
-
-// FormatExpiresAt encodes a deadline for MetaSessionExpiresAt. Always
-// emits UTC so a consumer in another timezone reads the same wall-clock
-// deadline.
-func FormatExpiresAt(t time.Time) string {
-	return t.UTC().Format(expiresAtLayout)
-}
-
-// ParseExpiresAt decodes the MetaSessionExpiresAt value back to a UTC
-// time. Returns an error when the field is empty or malformed; the
-// caller decides whether to fall back (zero value) or propagate.
-func ParseExpiresAt(s string) (time.Time, error) {
-	t, err := time.Parse(expiresAtLayout, s)
-	if err != nil {
-		return time.Time{}, err
-	}
-	return t.UTC(), nil
-}
-
-// FormatLeadMinutes encodes a lead duration for MetaSessionLeadMinutes
-// as the integer count of whole minutes. Sub-minute residuals are
-// truncated — the field is informational ("expires in ~N minutes") and
-// fractional minutes don't change what the UI displays.
-func FormatLeadMinutes(d time.Duration) string {
-	return strconv.Itoa(int(d / time.Minute))
-}
-
-// ParseLeadMinutes decodes a MetaSessionLeadMinutes value. Returns 0
-// and the parse error for malformed input; consumers that prefer a
-// silent fallback can simply ignore the error.
-func ParseLeadMinutes(s string) (int, error) {
-	return strconv.Atoi(s)
-}

client/internal/auth/sessionwatch/watcher.go

@@ -1,382 +0,0 @@
-// Package sessionwatch tracks the SSO session expiry deadline that the
-// management server publishes via LoginResponse / SyncResponse and fires
-// two warning events at fixed lead times before expiry: an interactive
-// T-WarningLead notification and a dismiss-gated T-FinalWarningLead
-// fallback dialog.
-//
-// The watcher is idempotent: Update may be called as often as the network
-// map snapshots arrive. Repeating the same deadline is a no-op; a new
-// deadline reschedules the timers and arms a fresh warning cycle.
-//
-// Warning firing is edge-detected. Each unique deadline value fires each
-// warning callback at most once.
-package sessionwatch
-
-import (
-	"errors"
-	"fmt"
-	"sync"
-	"time"
-
-	log "github.com/sirupsen/logrus"
-
-	cProto "github.com/netbirdio/netbird/client/proto"
-)
-
-const (
-	maxPastHorizon = 30 * 24 * time.Hour
-
-	// maxDeadlineHorizon caps how far in the future an accepted deadline
-	// can sit. A timestamp beyond this is almost certainly a protocol
-	// glitch, and silently arming a 100-year timer would hide the bug.
-	maxDeadlineHorizon = 10 * 365 * 24 * time.Hour
-
-	// WarningLead is how far before expiry the first (interactive)
-	// warning fires. Drives the T-10 OS notification with
-	// Extend/Dismiss actions.
-	WarningLead = 10 * time.Minute
-
-	// FinalWarningLead is how far before expiry the fallback final
-	// warning fires. Drives the auto-opened SessionAboutToExpire dialog,
-	// but only when the user has not dismissed the T-WarningLead warning
-	// for the same deadline. Must be strictly less than WarningLead.
-	FinalWarningLead = 2 * time.Minute
-)
-
-var (
-	// ErrDeadlineBeforeEpoch is returned by Update when the supplied
-	// deadline pre-dates 1970-01-01.
-	ErrDeadlineBeforeEpoch = errors.New("session deadline before unix epoch")
-
-	// ErrDeadlineTooFarFuture is returned by Update when the supplied
-	// deadline is more than maxDeadlineHorizon in the future.
-	ErrDeadlineTooFarFuture = errors.New("session deadline too far in the future")
-
-	// ErrDeadlineInPast is returned by Update when the supplied deadline
-	// is more than maxPastHorizon in the past.
-	ErrDeadlineInPast = errors.New("session deadline in the past")
-)
-
-// StatusRecorder is the side-effect surface the watcher drives on every
-// state transition. Production wires this to peer.Status (SetSessionExpiresAt
-// for deadline change/clear, PublishEvent for the two warnings); tests pass
-// a fake recorder so the same surface is observable without an engine.
-//
-// While the watcher runs, it owns the deadline propagated to the recorder:
-// every set, clear and sanity-check rejection routes the value through
-// SetSessionExpiresAt, so the SubscribeStatus snapshot the UI reads can
-// never drift from the watcher's timer state. (SetSessionExpiresAt fans
-// out its own state-change notification, so no separate notify is needed.)
-// The recorder is server-scoped and outlives this engine-scoped watcher;
-// Close deliberately leaves the recorder value in place so transient engine
-// restarts don't blank it — the client run loop clears it on real teardown.
-//
-// PublishEvent's signature mirrors peer.Status.PublishEvent: the watcher
-// composes the metadata internally so the wire format (MetaSession*) is
-// owned by sessionwatch, not the caller.
-type StatusRecorder interface {
-	SetSessionExpiresAt(deadline time.Time)
-	PublishEvent(
-		severity cProto.SystemEvent_Severity,
-		category cProto.SystemEvent_Category,
-		message string,
-		userMessage string,
-		metadata map[string]string,
-	)
-}
-
-// Watcher observes the latest session deadline and fires two warnings
-// before it expires: the interactive T-WarningLead notification, and the
-// fallback T-FinalWarningLead dialog (suppressed when the user dismissed
-// the first one for the same deadline). Safe for concurrent use.
-type Watcher struct {
-	lead      time.Duration
-	finalLead time.Duration
-
-	mu           sync.Mutex
-	current      time.Time
-	timer        *time.Timer
-	finalTimer   *time.Timer
-	firedAt      time.Time // deadline value the T-WarningLead callback last fired against
-	finalFiredAt time.Time // deadline value the T-FinalWarningLead callback last fired against
-	dismissedAt  time.Time // deadline value the user dismissed via Dismiss(); gates fireFinal
-	closed       bool
-	recorder     StatusRecorder
-}
-
-// New returns a watcher with the package defaults WarningLead and
-// FinalWarningLead. Pass nil for recorder to silence side effects (handy
-// in unit tests that exercise sanity checks without observing the publish
-// path).
-func New(recorder StatusRecorder) *Watcher {
-	return NewWithLeads(WarningLead, FinalWarningLead, recorder)
-}
-
-// NewWithLeads returns a watcher with custom lead times. Useful for tests.
-// final must be strictly less than lead; otherwise both timers fire in the
-// wrong order or simultaneously and the UI flow breaks. A zero final lead
-// disables the final-warning timer entirely (see armTimerLocked) so a
-// millisecond-scale deadline doesn't flush both timers in one tick.
-func NewWithLeads(lead, final time.Duration, recorder StatusRecorder) *Watcher {
-	return &Watcher{
-		lead:      lead,
-		finalLead: final,
-		recorder:  recorder,
-	}
-}
-
-// Update sets the latest deadline. Pass the zero time to clear (e.g. when
-// a Sync push from the server omits the field because login expiration
-// was disabled).
-//
-// Same-value updates are no-ops. A different non-zero value cancels any
-// pending timer, resets the "already fired" guards, and — when the
-// deadline lies in the future — arms fresh warning timers. A deadline
-// already in the past (within maxPastHorizon) is recorded as-is with no
-// timers: the session has expired and consumers render it that way.
-//
-// Returns one of the sentinel Err* values when the deadline fails the
-// sanity checks (pre-epoch, far future, or past beyond maxPastHorizon).
-// In every error case the watcher first clears its state so it stays
-// consistent with what the caller will push into its other sinks (e.g.
-// applySessionDeadline forces a zero deadline into the status recorder
-// after a non-nil error).
-func (w *Watcher) Update(deadline time.Time) error {
-	w.mu.Lock()
-	if w.closed {
-		w.mu.Unlock()
-		return nil
-	}
-
-	if deadline.IsZero() {
-		w.clearLocked()
-		return nil
-	}
-
-	now := time.Now()
-	switch {
-	case deadline.Before(time.Unix(0, 0)):
-		w.clearLocked()
-		return fmt.Errorf("%w: %v", ErrDeadlineBeforeEpoch, deadline)
-	case deadline.After(now.Add(maxDeadlineHorizon)):
-		w.clearLocked()
-		return fmt.Errorf("%w: %v", ErrDeadlineTooFarFuture, deadline)
-	case deadline.Before(now.Add(-maxPastHorizon)):
-		w.clearLocked()
-		return fmt.Errorf("%w: %v (now=%v)", ErrDeadlineInPast, deadline, now)
-	}
-
-	if deadline.Equal(w.current) {
-		w.mu.Unlock()
-		return nil
-	}
-
-	w.stopTimerLocked()
-	w.current = deadline
-	// Reset every per-deadline guard so a refreshed deadline arms a fresh
-	// warning cycle: both edge triggers and the user Dismiss decision
-	// (the user agreed to the old deadline expiring; a new deadline
-	// restarts the contract).
-	w.firedAt = time.Time{}
-	w.finalFiredAt = time.Time{}
-	w.dismissedAt = time.Time{}
-
-	if deadline.After(now) {
-		w.armTimerLocked(deadline)
-	}
-	recorder := w.recorder
-	w.mu.Unlock()
-	if recorder != nil {
-		recorder.SetSessionExpiresAt(deadline)
-	}
-	log.Infof("auth session deadline set to: %s (in %s)", deadline.Format(time.RFC3339), time.Until(deadline).Round(time.Second))
-	return nil
-}
-
-// Deadline returns the most recently observed deadline. Zero when no
-// deadline is currently tracked.
-func (w *Watcher) Deadline() time.Time {
-	w.mu.Lock()
-	defer w.mu.Unlock()
-	return w.current
-}
-
-// Dismiss records the user's "Dismiss" action against the current deadline
-// and suppresses the upcoming final-warning callback for that deadline.
-// Idempotent: repeated calls are no-ops. A subsequent Update with a fresh
-// deadline resets the dismissal so the final-warning cycle re-arms.
-//
-// No-op when the watcher holds no deadline or has been closed.
-func (w *Watcher) Dismiss() {
-	w.mu.Lock()
-	defer w.mu.Unlock()
-	if w.closed || w.current.IsZero() {
-		return
-	}
-	if w.dismissedAt.Equal(w.current) {
-		return
-	}
-	w.dismissedAt = w.current
-	// Cancel the armed final-warning timer eagerly. fireFinal would also
-	// gate on dismissedAt, but stopping the timer avoids a wakeup with
-	// nothing to do and makes the intent visible.
-	if w.finalTimer != nil {
-		w.finalTimer.Stop()
-		w.finalTimer = nil
-	}
-	log.Infof("auth session final-warning dismissed for deadline %s", w.current.Format(time.RFC3339))
-}
-
-// Close stops any pending timer. Update calls after Close are ignored.
-// The recorder keeps its deadline: the watcher is engine-scoped and closes
-// on every engine restart (network change, sleep/wake, stream errors)
-// while the SSO deadline stays valid across those, so clearing here would
-// blank the UI's "expires in" row on every transient reconnect. The
-// client run loop clears the server-scoped recorder when it exits for
-// real (Down, profile switch, permanent login failure).
-func (w *Watcher) Close() {
-	w.mu.Lock()
-	defer w.mu.Unlock()
-	if w.closed {
-		return
-	}
-	w.closed = true
-	w.stopTimerLocked()
-	w.current = time.Time{}
-	w.firedAt = time.Time{}
-	w.finalFiredAt = time.Time{}
-	w.dismissedAt = time.Time{}
-}
-
-// clearLocked drops the tracked deadline and notifies the recorder so
-// downstream consumers (SubscribeStatus stream, UI) drop their anchor.
-// The caller must hold w.mu; this helper releases it before invoking
-// the recorder.
-func (w *Watcher) clearLocked() {
-	if w.current.IsZero() {
-		w.mu.Unlock()
-		return
-	}
-	w.stopTimerLocked()
-	w.current = time.Time{}
-	w.firedAt = time.Time{}
-	w.finalFiredAt = time.Time{}
-	w.dismissedAt = time.Time{}
-	recorder := w.recorder
-	w.mu.Unlock()
-	if recorder != nil {
-		recorder.SetSessionExpiresAt(time.Time{})
-	}
-	log.Infof("auth session deadline cleared")
-}
-
-func (w *Watcher) stopTimerLocked() {
-	if w.timer != nil {
-		w.timer.Stop()
-		w.timer = nil
-	}
-	if w.finalTimer != nil {
-		w.finalTimer.Stop()
-		w.finalTimer = nil
-	}
-}
-
-func (w *Watcher) armTimerLocked(deadline time.Time) {
-	w.timer = armOneShotLocked(deadline.Add(-w.lead), func() { w.fire(deadline) })
-	// finalLead <= 0 disables the final-warning timer entirely. Used by
-	// tests that predate the final-warning fallback so a millisecond-scale
-	// deadline does not flush both timers at once.
-	if w.finalLead > 0 {
-		w.finalTimer = armOneShotLocked(deadline.Add(-w.finalLead), func() { w.fireFinal(deadline) })
-	}
-}
-
-func (w *Watcher) fire(armedFor time.Time) {
-	w.mu.Lock()
-	if w.closed || !w.current.Equal(armedFor) {
-		// Deadline moved while we were waiting (e.g. a successful extend).
-		// The reschedule path armed a fresh timer; this one is stale.
-		w.mu.Unlock()
-		return
-	}
-	if !w.firedAt.IsZero() && w.firedAt.Equal(armedFor) {
-		w.mu.Unlock()
-		return
-	}
-	w.firedAt = armedFor
-	recorder := w.recorder
-	w.mu.Unlock()
-	if recorder == nil {
-		return
-	}
-	log.Infof("auth session expiry soon warning fired")
-	publishWarning(recorder, armedFor, false)
-}
-
-// fireFinal mirrors fire for the T-FinalWarningLead timer with an extra
-// dismiss-gate: if the user dismissed the T-WarningLead notification for
-// this deadline, the final warning is suppressed entirely.
-func (w *Watcher) fireFinal(armedFor time.Time) {
-	w.mu.Lock()
-	if w.closed || !w.current.Equal(armedFor) {
-		w.mu.Unlock()
-		return
-	}
-	if !w.finalFiredAt.IsZero() && w.finalFiredAt.Equal(armedFor) {
-		w.mu.Unlock()
-		return
-	}
-	if w.dismissedAt.Equal(armedFor) {
-		w.mu.Unlock()
-		log.Infof("auth session final-warning skipped (dismissed by user)")
-		return
-	}
-	w.finalFiredAt = armedFor
-	recorder := w.recorder
-	w.mu.Unlock()
-	if recorder == nil {
-		return
-	}
-	log.Infof("auth session final-warning fired")
-	publishWarning(recorder, armedFor, true)
-}
-
-// armOneShotLocked schedules cb at fireAt. When fireAt is already in the
-// past it dispatches on the next scheduler tick so a state-change recorder
-// notification (invoked after w.mu is released) lands first. Caller must
-// hold w.mu.
-func armOneShotLocked(fireAt time.Time, cb func()) *time.Timer {
-	delay := time.Until(fireAt)
-	if delay <= 0 {
-		return time.AfterFunc(0, cb)
-	}
-	return time.AfterFunc(delay, cb)
-}
-
-// publishWarning composes the SystemEvent for a watcher-fired warning and
-// pushes it through the recorder. Severity is CRITICAL on both — bypassing
-// the user's Notifications toggle is deliberate: missing the warning
-// window forces the post-mortem SessionExpired flow (tunnel torn down,
-// lock icon, manual re-login), which is the UX we are trying to avoid.
-func publishWarning(recorder StatusRecorder, deadline time.Time, final bool) {
-	lead := WarningLead
-	message := "session expiry warning"
-	meta := map[string]string{
-		MetaSessionWarning:   "true",
-		MetaSessionExpiresAt: FormatExpiresAt(deadline),
-	}
-	if final {
-		lead = FinalWarningLead
-		message = "session expiry final warning"
-		meta[MetaSessionFinal] = "true"
-	}
-	meta[MetaSessionLeadMinutes] = FormatLeadMinutes(lead)
-
-	recorder.PublishEvent(
-		cProto.SystemEvent_CRITICAL,
-		cProto.SystemEvent_AUTHENTICATION,
-		message,
-		"",
-		meta,
-	)
-}

client/internal/auth/sessionwatch/watcher_test.go

@@ -1,529 +0,0 @@
-package sessionwatch
-
-import (
-	"errors"
-	"sync"
-	"testing"
-	"time"
-
-	cProto "github.com/netbirdio/netbird/client/proto"
-)
-
-// fakeRecorder satisfies StatusRecorder and records every call so tests
-// can observe what the watcher emits. SetSessionExpiresAt and PublishEvent
-// land in the same ordered events slice (with the Kind distinguishing
-// them) so tests that care about ordering still work. lastDeadline holds
-// the most recent value passed to SetSessionExpiresAt so tests can assert
-// the recorder ended up cleared/set as expected.
-type fakeRecorder struct {
-	mu           sync.Mutex
-	events       []event
-	lastDeadline time.Time
-}
-
-type eventKind int
-
-const (
-	stateChange eventKind = iota
-	publish
-)
-
-type event struct {
-	kind eventKind
-	// Set only for publish events.
-	severity cProto.SystemEvent_Severity
-	category cProto.SystemEvent_Category
-	message  string
-	meta     map[string]string
-}
-
-// SetSessionExpiresAt mirrors peer.Status: a same-value write is a no-op,
-// a real change records the new value and fans out a state-change (the
-// production recorder calls notifyStateChange internally). The baseline
-// is the zero time, so an initial clear before any deadline is set emits
-// nothing — matching the real recorder.
-func (r *fakeRecorder) SetSessionExpiresAt(deadline time.Time) {
-	r.mu.Lock()
-	defer r.mu.Unlock()
-	if r.lastDeadline.Equal(deadline) {
-		return
-	}
-	r.lastDeadline = deadline
-	r.events = append(r.events, event{kind: stateChange})
-}
-
-func (r *fakeRecorder) deadline() time.Time {
-	r.mu.Lock()
-	defer r.mu.Unlock()
-	return r.lastDeadline
-}
-
-func (r *fakeRecorder) PublishEvent(
-	severity cProto.SystemEvent_Severity,
-	category cProto.SystemEvent_Category,
-	message string,
-	_ string,
-	metadata map[string]string,
-) {
-	r.mu.Lock()
-	defer r.mu.Unlock()
-	r.events = append(r.events, event{
-		kind:     publish,
-		severity: severity,
-		category: category,
-		message:  message,
-		meta:     metadata,
-	})
-}
-
-func (r *fakeRecorder) snapshot() []event {
-	r.mu.Lock()
-	defer r.mu.Unlock()
-	out := make([]event, len(r.events))
-	copy(out, r.events)
-	return out
-}
-
-func (e event) isFinalWarning() bool {
-	return e.kind == publish && e.meta[MetaSessionFinal] == "true"
-}
-
-func (e event) isWarning() bool {
-	return e.kind == publish && e.meta[MetaSessionWarning] == "true" && e.meta[MetaSessionFinal] != "true"
-}
-
-func countWhere(events []event, pred func(event) bool) int {
-	n := 0
-	for _, e := range events {
-		if pred(e) {
-			n++
-		}
-	}
-	return n
-}
-
-func waitForEvents(t *testing.T, r *fakeRecorder, want int) []event {
-	t.Helper()
-	deadline := time.Now().Add(500 * time.Millisecond)
-	for time.Now().Before(deadline) {
-		if got := r.snapshot(); len(got) >= want {
-			return got
-		}
-		time.Sleep(5 * time.Millisecond)
-	}
-	got := r.snapshot()
-	t.Fatalf("timed out waiting for %d events, got %d: %+v", want, len(got), got)
-	return nil
-}
-
-// newWatcher builds a watcher with the final timer disabled (finalLead=0),
-// matching the lead-only behaviour the pre-final-warning tests assume.
-func newWatcher(lead time.Duration, r *fakeRecorder) *Watcher {
-	return NewWithLeads(lead, 0, r)
-}
-
-func TestUpdateZeroBeforeAnythingIsNoop(t *testing.T) {
-	r := &fakeRecorder{}
-	w := newWatcher(50*time.Millisecond, r)
-	defer w.Close()
-
-	_ = w.Update(time.Time{})
-
-	if got := r.snapshot(); len(got) != 0 {
-		t.Fatalf("expected no events on initial zero, got %+v", got)
-	}
-}
-
-func TestUpdateNonZeroFiresStateChange(t *testing.T) {
-	r := &fakeRecorder{}
-	w := newWatcher(50*time.Millisecond, r)
-	defer w.Close()
-
-	d := time.Now().Add(time.Hour)
-	_ = w.Update(d)
-
-	events := waitForEvents(t, r, 1)
-	if events[0].kind != stateChange {
-		t.Fatalf("expected stateChange, got %+v", events[0])
-	}
-	if !w.Deadline().Equal(d) {
-		t.Fatalf("deadline mismatch: %v vs %v", w.Deadline(), d)
-	}
-}
-
-func TestSameDeadlineIsNoop(t *testing.T) {
-	r := &fakeRecorder{}
-	w := newWatcher(50*time.Millisecond, r)
-	defer w.Close()
-
-	d := time.Now().Add(time.Hour)
-	_ = w.Update(d)
-	_ = w.Update(d)
-	_ = w.Update(d)
-
-	events := waitForEvents(t, r, 1)
-	if len(events) != 1 {
-		t.Fatalf("expected exactly 1 event for repeated same deadline, got %d: %+v", len(events), events)
-	}
-}
-
-func TestWarningFiresOnceWithinLeadWindow(t *testing.T) {
-	r := &fakeRecorder{}
-	lead := 50 * time.Millisecond
-	w := newWatcher(lead, r)
-	defer w.Close()
-
-	// Deadline 80ms out — warning should fire after ~30ms.
-	d := time.Now().Add(80 * time.Millisecond)
-	_ = w.Update(d)
-
-	events := waitForEvents(t, r, 2)
-	if events[0].kind != stateChange {
-		t.Fatalf("event[0] should be stateChange, got %+v", events[0])
-	}
-	if !events[1].isWarning() {
-		t.Fatalf("event[1] should be a warning publish, got %+v", events[1])
-	}
-}
-
-func TestWarningFiresImmediatelyWhenAlreadyInsideWindow(t *testing.T) {
-	r := &fakeRecorder{}
-	w := newWatcher(time.Hour, r) // lead > delta => fire immediately
-	defer w.Close()
-
-	d := time.Now().Add(10 * time.Millisecond)
-	_ = w.Update(d)
-
-	events := waitForEvents(t, r, 2)
-	if !events[1].isWarning() {
-		t.Fatalf("expected immediate warning publish, got %+v", events[1])
-	}
-}
-
-func TestNewDeadlineCancelsPriorTimer(t *testing.T) {
-	r := &fakeRecorder{}
-	lead := 50 * time.Millisecond
-	w := newWatcher(lead, r)
-	defer w.Close()
-
-	first := time.Now().Add(80 * time.Millisecond) // would fire warning ~30ms in
-	_ = w.Update(first)
-
-	// Replace with a far-future deadline before the warning fires.
-	time.Sleep(5 * time.Millisecond)
-	second := time.Now().Add(time.Hour)
-	_ = w.Update(second)
-
-	// Wait past when first's warning would have fired.
-	time.Sleep(80 * time.Millisecond)
-
-	if n := countWhere(r.snapshot(), event.isWarning); n != 0 {
-		t.Fatalf("warning fired for cancelled deadline: %+v", r.snapshot())
-	}
-}
-
-func TestRefreshAfterFireArmsNewWarning(t *testing.T) {
-	r := &fakeRecorder{}
-	lead := 150 * time.Millisecond
-	w := newWatcher(lead, r)
-	defer w.Close()
-
-	// Warning fires ~20ms in; the deadline itself stays 150ms away so the
-	// replacement below lands well before it.
-	first := time.Now().Add(170 * time.Millisecond)
-	_ = w.Update(first)
-
-	// Wait for stateChange + warning of the first cycle.
-	waitForEvents(t, r, 2)
-
-	// Simulate a successful extend: brand new deadline.
-	second := time.Now().Add(60 * time.Millisecond)
-	_ = w.Update(second)
-
-	// 4 events total: stateChange, warning (first), stateChange, warning (second).
-	events := waitForEvents(t, r, 4)
-	if events[2].kind != stateChange {
-		t.Fatalf("event[2] should be stateChange for the new deadline, got %+v", events[2])
-	}
-	if !events[3].isWarning() {
-		t.Fatalf("event[3] should be a warning publish for the new deadline, got %+v", events[3])
-	}
-}
-
-func TestUpdateZeroAfterNonZeroClearsState(t *testing.T) {
-	r := &fakeRecorder{}
-	w := newWatcher(time.Hour, r)
-	defer w.Close()
-
-	d := time.Now().Add(2 * time.Hour)
-	_ = w.Update(d)
-	waitForEvents(t, r, 1)
-
-	_ = w.Update(time.Time{})
-
-	events := waitForEvents(t, r, 2)
-	if events[1].kind != stateChange {
-		t.Fatalf("expected stateChange on clear, got %+v", events[1])
-	}
-	if !w.Deadline().IsZero() {
-		t.Fatalf("Deadline should be zero after clear")
-	}
-}
-
-func TestUpdateRejectsBeforeEpoch(t *testing.T) {
-	r := &fakeRecorder{}
-	w := newWatcher(50*time.Millisecond, r)
-	defer w.Close()
-
-	good := time.Now().Add(time.Hour)
-	if err := w.Update(good); err != nil {
-		t.Fatalf("seed Update: %v", err)
-	}
-
-	err := w.Update(time.Unix(-100, 0))
-	if !errors.Is(err, ErrDeadlineBeforeEpoch) {
-		t.Fatalf("want ErrDeadlineBeforeEpoch, got %v", err)
-	}
-	if !w.Deadline().IsZero() {
-		t.Fatalf("rejected pre-epoch update must clear deadline; got %v", w.Deadline())
-	}
-}
-
-func TestUpdateRejectsTooFarFuture(t *testing.T) {
-	r := &fakeRecorder{}
-	w := newWatcher(50*time.Millisecond, r)
-	defer w.Close()
-
-	good := time.Now().Add(time.Hour)
-	if err := w.Update(good); err != nil {
-		t.Fatalf("seed Update: %v", err)
-	}
-
-	err := w.Update(time.Now().Add(50 * 365 * 24 * time.Hour))
-	if !errors.Is(err, ErrDeadlineTooFarFuture) {
-		t.Fatalf("want ErrDeadlineTooFarFuture, got %v", err)
-	}
-	if !w.Deadline().IsZero() {
-		t.Fatalf("rejected far-future update must clear deadline; got %v", w.Deadline())
-	}
-}
-
-func TestUpdateRecentPastRecordedAsExpired(t *testing.T) {
-	r := &fakeRecorder{}
-	w := newWatcher(50*time.Millisecond, r)
-	defer w.Close()
-
-	d := time.Now().Add(-1 * time.Hour)
-	if err := w.Update(d); err != nil {
-		t.Fatalf("recent-past Update should succeed, got %v", err)
-	}
-	if !w.Deadline().Equal(d) {
-		t.Fatalf("expected deadline to be recorded, got %v want %v", w.Deadline(), d)
-	}
-	if got := r.deadline(); !got.Equal(d) {
-		t.Fatalf("recorder deadline = %v, want %v", got, d)
-	}
-
-	time.Sleep(80 * time.Millisecond)
-	if n := countWhere(r.snapshot(), func(e event) bool { return e.kind == publish }); n != 0 {
-		t.Fatalf("no warning events may fire for an already-past deadline, got %+v", r.snapshot())
-	}
-}
-
-func TestUpdateAncientPastRejected(t *testing.T) {
-	r := &fakeRecorder{}
-	w := newWatcher(50*time.Millisecond, r)
-	defer w.Close()
-
-	good := time.Now().Add(time.Hour)
-	if err := w.Update(good); err != nil {
-		t.Fatalf("seed Update: %v", err)
-	}
-	// Drain the stateChange from the seed.
-	waitForEvents(t, r, 1)
-
-	err := w.Update(time.Now().Add(-31 * 24 * time.Hour))
-	if !errors.Is(err, ErrDeadlineInPast) {
-		t.Fatalf("want ErrDeadlineInPast, got %v", err)
-	}
-	if !w.Deadline().IsZero() {
-		t.Fatalf("rejected ancient-past update must clear the deadline, got %v", w.Deadline())
-	}
-	events := waitForEvents(t, r, 2)
-	if events[1].kind != stateChange {
-		t.Fatalf("expected stateChange on clear, got %+v", events[1])
-	}
-}
-
-func TestCloseSilencesUpdates(t *testing.T) {
-	r := &fakeRecorder{}
-	w := newWatcher(50*time.Millisecond, r)
-	w.Close()
-
-	if err := w.Update(time.Now().Add(time.Hour)); err != nil {
-		t.Fatalf("Update after Close: want nil, got %v", err)
-	}
-	if got := r.snapshot(); len(got) != 0 {
-		t.Fatalf("expected no events after Close, got %+v", got)
-	}
-}
-
-// TestCloseKeepsRecorderDeadline pins the reconnect-flap fix: the watcher
-// closes on every engine restart (network change, sleep/wake) while the
-// SSO deadline stays valid across those, so Close must leave the
-// server-scoped recorder's value in place. The client run loop clears the
-// recorder when it exits for real.
-func TestCloseKeepsRecorderDeadline(t *testing.T) {
-	r := &fakeRecorder{}
-	w := newWatcher(time.Hour, r)
-
-	d := time.Now().Add(2 * time.Hour)
-	if err := w.Update(d); err != nil {
-		t.Fatalf("seed Update: %v", err)
-	}
-	if got := r.deadline(); !got.Equal(d) {
-		t.Fatalf("recorder deadline after Update = %v, want %v", got, d)
-	}
-
-	w.Close()
-
-	if got := r.deadline(); !got.Equal(d) {
-		t.Fatalf("recorder deadline after Close = %v, want %v", got, d)
-	}
-}
-
-// TestCloseWithoutDeadlineLeavesRecorderUntouched guards the symmetric
-// case: closing a watcher that never held a deadline must not emit a
-// redundant clear (the recorder may legitimately hold a value written by
-// some other path; the watcher only owns what it set).
-func TestCloseWithoutDeadlineLeavesRecorderUntouched(t *testing.T) {
-	r := &fakeRecorder{}
-	w := newWatcher(time.Hour, r)
-
-	w.Close()
-
-	if got := r.snapshot(); len(got) != 0 {
-		t.Fatalf("expected no events from Close on an empty watcher, got %+v", got)
-	}
-}
-
-func TestFinalWarningFiresAfterRegularWarning(t *testing.T) {
-	r := &fakeRecorder{}
-	// Warning fires at deadline-80ms, final at deadline-30ms.
-	w := NewWithLeads(80*time.Millisecond, 30*time.Millisecond, r)
-	defer w.Close()
-
-	d := time.Now().Add(100 * time.Millisecond)
-	_ = w.Update(d)
-
-	// Expect stateChange + warning + final-warning.
-	events := waitForEvents(t, r, 3)
-
-	if countWhere(events, func(e event) bool { return e.kind == stateChange }) != 1 {
-		t.Fatalf("expected exactly 1 stateChange, got %+v", events)
-	}
-	if countWhere(events, event.isWarning) != 1 {
-		t.Fatalf("expected exactly 1 warning publish, got %+v", events)
-	}
-	if countWhere(events, event.isFinalWarning) != 1 {
-		t.Fatalf("expected exactly 1 final-warning publish, got %+v", events)
-	}
-
-	// Warning must precede final (same deadline, longer lead fires first).
-	var wIdx, fIdx int
-	for i, e := range events {
-		switch {
-		case e.isWarning():
-			wIdx = i
-		case e.isFinalWarning():
-			fIdx = i
-		}
-	}
-	if wIdx > fIdx {
-		t.Fatalf("warning must publish before final-warning, got order %+v", events)
-	}
-}
-
-func TestDismissSuppressesFinalWarning(t *testing.T) {
-	r := &fakeRecorder{}
-	w := NewWithLeads(80*time.Millisecond, 30*time.Millisecond, r)
-	defer w.Close()
-
-	d := time.Now().Add(100 * time.Millisecond)
-	_ = w.Update(d)
-
-	// Wait for the warning publish so we know we're inside the warning
-	// window, then dismiss before the final timer would fire.
-	deadline := time.Now().Add(500 * time.Millisecond)
-	for time.Now().Before(deadline) {
-		if countWhere(r.snapshot(), event.isWarning) >= 1 {
-			break
-		}
-		time.Sleep(2 * time.Millisecond)
-	}
-	if countWhere(r.snapshot(), event.isWarning) < 1 {
-		t.Fatalf("warning did not publish in time, events=%+v", r.snapshot())
-	}
-
-	w.Dismiss()
-
-	// Now wait past when the final would have fired.
-	time.Sleep(120 * time.Millisecond)
-
-	if n := countWhere(r.snapshot(), event.isFinalWarning); n != 0 {
-		t.Fatalf("final-warning published after Dismiss(), events=%+v", r.snapshot())
-	}
-}
-
-func TestDismissResetByNewDeadline(t *testing.T) {
-	r := &fakeRecorder{}
-	w := NewWithLeads(80*time.Millisecond, 30*time.Millisecond, r)
-	defer w.Close()
-
-	first := time.Now().Add(100 * time.Millisecond)
-	_ = w.Update(first)
-
-	// Dismiss against the first deadline.
-	w.Dismiss()
-
-	// Replace with a fresh deadline before the first's timers complete.
-	time.Sleep(10 * time.Millisecond)
-	second := time.Now().Add(100 * time.Millisecond)
-	_ = w.Update(second)
-
-	// The second cycle must publish a final-warning (the dismiss state
-	// did not carry over).
-	deadline := time.Now().Add(500 * time.Millisecond)
-	for time.Now().Before(deadline) {
-		if countWhere(r.snapshot(), event.isFinalWarning) >= 1 {
-			break
-		}
-		time.Sleep(5 * time.Millisecond)
-	}
-	if countWhere(r.snapshot(), event.isFinalWarning) < 1 {
-		t.Fatalf("final-warning did not publish on fresh deadline after Dismiss reset, events=%+v", r.snapshot())
-	}
-}
-
-func TestDismissBeforeUpdateIsNoop(t *testing.T) {
-	r := &fakeRecorder{}
-	w := NewWithLeads(80*time.Millisecond, 30*time.Millisecond, r)
-	defer w.Close()
-
-	// No deadline tracked yet; Dismiss must be a no-op (no panic, no state).
-	w.Dismiss()
-
-	d := time.Now().Add(100 * time.Millisecond)
-	_ = w.Update(d)
-
-	// Final warning should still publish — Dismiss only acts on the current
-	// deadline, and there was none at the time of the call.
-	deadline := time.Now().Add(500 * time.Millisecond)
-	for time.Now().Before(deadline) {
-		if countWhere(r.snapshot(), event.isFinalWarning) >= 1 {
-			return
-		}
-		time.Sleep(5 * time.Millisecond)
-	}
-	t.Fatalf("final-warning did not publish after no-op pre-Update Dismiss, events=%+v", r.snapshot())
-}

client/internal/auth/util.go

@@ -20,26 +20,14 @@ func randomBytesInHex(count int) (string, error) {
 	return hex.EncodeToString(buf), nil
 }
 
-// validateTokenAudience checks that the token is a well-formed JWT whose
+// isValidAccessToken is a simple validation of the access token
-// audience claim matches the expected audience.
+func isValidAccessToken(token string, audience string) error {
-//
-// It does NOT verify the token's cryptographic signature and therefore must not
-// be treated as an authenticity check. The token is obtained by the client
-// directly from the IdP token endpoint over TLS, and its signature is verified
-// server-side by the management server against the IdP's JWKS
-// (see shared/auth/jwt/validator.go). This function is only a client-side
-// sanity check that the returned token targets the expected audience.
-func validateTokenAudience(token string, audience string) error {
 	if token == "" {
 		return fmt.Errorf("token received is empty")
 	}
 
-	parts := strings.Split(token, ".")
+	encodedClaims := strings.Split(token, ".")[1]
-	if len(parts) != 3 {
+	claimsString, err := base64.RawURLEncoding.DecodeString(encodedClaims)
-		return fmt.Errorf("token is not a well-formed JWT")
-	}
-
-	claimsString, err := base64.RawURLEncoding.DecodeString(parts[1])
 	if err != nil {
 		return err
 	}

client/internal/auth/util_test.go

@@ -1,108 +0,0 @@
-package auth
-
-import (
-	"encoding/base64"
-	"encoding/json"
-	"testing"
-)
-
-// makeJWT builds an unsigned JWT-shaped string (header.payload.signature) with
-// the given claims payload. The signature part is arbitrary because
-// validateTokenAudience intentionally does not verify it.
-func makeJWT(t *testing.T, claims map[string]interface{}) string {
-	t.Helper()
-	header := base64.RawURLEncoding.EncodeToString([]byte(`{"alg":"RS256","typ":"JWT"}`))
-	payloadBytes, err := json.Marshal(claims)
-	if err != nil {
-		t.Fatalf("marshal claims: %v", err)
-	}
-	payload := base64.RawURLEncoding.EncodeToString(payloadBytes)
-	return header + "." + payload + ".unverified-signature"
-}
-
-func TestValidateTokenAudience(t *testing.T) {
-	tests := []struct {
-		name     string
-		token    string
-		audience string
-		wantErr  bool
-	}{
-		{
-			name:     "empty token",
-			token:    "",
-			audience: "netbird",
-			wantErr:  true,
-		},
-		{
-			name:     "not a JWT - no dots",
-			token:    "notajwt",
-			audience: "netbird",
-			wantErr:  true,
-		},
-		{
-			name:     "not a JWT - two parts only",
-			token:    "header.payload",
-			audience: "netbird",
-			wantErr:  true,
-		},
-		{
-			name:     "matching string audience",
-			token:    makeJWT(t, map[string]interface{}{"aud": "netbird"}),
-			audience: "netbird",
-			wantErr:  false,
-		},
-		{
-			name:     "mismatching string audience",
-			token:    makeJWT(t, map[string]interface{}{"aud": "other"}),
-			audience: "netbird",
-			wantErr:  true,
-		},
-		{
-			name:     "matching audience in array",
-			token:    makeJWT(t, map[string]interface{}{"aud": []interface{}{"other", "netbird"}}),
-			audience: "netbird",
-			wantErr:  false,
-		},
-		{
-			name:     "mismatching audience array",
-			token:    makeJWT(t, map[string]interface{}{"aud": []interface{}{"a", "b"}}),
-			audience: "netbird",
-			wantErr:  true,
-		},
-		{
-			name:     "missing audience claim",
-			token:    makeJWT(t, map[string]interface{}{"sub": "user"}),
-			audience: "netbird",
-			wantErr:  true,
-		},
-		{
-			name:     "invalid base64 payload",
-			token:    "header.!!!not-base64!!!.sig",
-			audience: "netbird",
-			wantErr:  true,
-		},
-	}
-
-	for _, tc := range tests {
-		t.Run(tc.name, func(t *testing.T) {
-			err := validateTokenAudience(tc.token, tc.audience)
-			if tc.wantErr && err == nil {
-				t.Fatalf("expected error, got nil")
-			}
-			if !tc.wantErr && err != nil {
-				t.Fatalf("expected no error, got %v", err)
-			}
-		})
-	}
-}
-
-// TestValidateTokenAudienceNoPanic guards the regression where a non-empty
-// token without the JWT dot structure caused an index-out-of-range panic.
-func TestValidateTokenAudienceNoPanic(t *testing.T) {
-	inputs := []string{"a", ".", "a.", "aaaa", "no-dots-here"}
-	for _, in := range inputs {
-		if err := validateTokenAudience(in, "netbird"); err == nil {
-			t.Fatalf("expected error for malformed token %q, got nil", in)
-		}
-	}
-}

client/internal/conn_mgr.go

@@ -34,8 +34,6 @@ const (
 // - Handling connection establishment based on peer signaling
 //
 // The implementation is not thread-safe; it is protected by engine.syncMsgMux.
-// The only exception is ActivatePeer, which is safe for concurrent use so the
-// DNS warm-up path can call it without contending on the engine mutex.
 type ConnMgr struct {
 	peerStore        *peerstore.Store
 	statusRecorder   *peer.Status
@@ -44,26 +42,12 @@ type ConnMgr struct {
 	rosenpassEnabled bool
 
 	lazyConnMgr *manager.Manager
-	// lazyConnMgrMu guards the lazyConnMgr pointer for readers outside the
-	// engine loop (ActivatePeer). Writers hold it in addition to
-	// engine.syncMsgMux; all other reads stay under engine.syncMsgMux only.
-	lazyConnMgrMu sync.RWMutex
-
-	// reconcileRoutedIPs re-applies a peer's routed allowed IPs after its lazy wake endpoint is
-	// (re)armed (Mode A at arm time). Injected by the engine; nil disables the reconcile.
-	reconcileRoutedIPs func(peerKey string) error
 
 	wg            sync.WaitGroup
 	lazyCtx       context.Context
 	lazyCtxCancel context.CancelFunc
 }
 
-// SetRoutedIPsReconciler injects the callback used to re-apply a peer's routed allowed IPs when
-// its lazy wake endpoint is (re)armed. Must be called before the lazy manager starts.
-func (e *ConnMgr) SetRoutedIPsReconciler(fn func(peerKey string) error) {
-	e.reconcileRoutedIPs = fn
-}
-
 func NewConnMgr(engineConfig *EngineConfig, statusRecorder *peer.Status, peerStore *peerstore.Store, iface lazyconn.WGIface) *ConnMgr {
 	e := &ConnMgr{
 		peerStore:        peerStore,
@@ -125,7 +109,7 @@ func (e *ConnMgr) UpdatedRemoteFeatureFlag(ctx context.Context, enabled bool) er
 			return nil
 		}
 
-		log.Infof("lazy connection manager is enabled by the management feature flag")
+		log.Warnf("lazy connection manager is enabled by management feature flag")
 		e.initLazyManager(ctx)
 		e.statusRecorder.UpdateLazyConnection(true)
 		return e.addPeersToLazyConnManager()
@@ -254,20 +238,12 @@ func (e *ConnMgr) RemovePeerConn(peerKey string) {
 	conn.Log.Infof("removed peer from lazy conn manager")
 }
 
-// ActivatePeer wakes an idle lazy connection. Unlike the rest of ConnMgr it is
-// safe for concurrent use: the lazy manager pointer is read under lazyConnMgrMu
-// and the manager itself is internally synchronized, so callers outside the
-// engine loop (DNS warm-up) do not need engine.syncMsgMux.
 func (e *ConnMgr) ActivatePeer(ctx context.Context, conn *peer.Conn) {
-	e.lazyConnMgrMu.RLock()
+	if !e.isStartedWithLazyMgr() {
-	lazyConnMgr := e.lazyConnMgr
-	started := lazyConnMgr != nil && e.lazyCtxCancel != nil
-	e.lazyConnMgrMu.RUnlock()
-	if !started {
 		return
 	}
 
-	if found := lazyConnMgr.ActivatePeer(conn.GetKey()); found {
+	if found := e.lazyConnMgr.ActivatePeer(conn.GetKey()); found {
 		if err := conn.Open(ctx); err != nil {
 			conn.Log.Errorf("failed to open connection: %v", err)
 		}
@@ -292,22 +268,16 @@ func (e *ConnMgr) Close() {
 
 	e.lazyCtxCancel()
 	e.wg.Wait()
-
-	e.lazyConnMgrMu.Lock()
 	e.lazyConnMgr = nil
-	e.lazyConnMgrMu.Unlock()
 }
 
 func (e *ConnMgr) initLazyManager(engineCtx context.Context) {
 	cfg := manager.Config{
 		InactivityThreshold: inactivityThresholdEnv(),
-		ReconcileAllowedIPs: e.reconcileRoutedIPs,
 	}
-
-	e.lazyConnMgrMu.Lock()
 	e.lazyConnMgr = manager.NewManager(cfg, engineCtx, e.peerStore, e.iface)
+
 	e.lazyCtx, e.lazyCtxCancel = context.WithCancel(engineCtx)
-	e.lazyConnMgrMu.Unlock()
 
 	e.wg.Add(1)
 	go func() {
@@ -346,10 +316,7 @@ func (e *ConnMgr) closeManager(ctx context.Context) {
 
 	e.lazyCtxCancel()
 	e.wg.Wait()
-
-	e.lazyConnMgrMu.Lock()
 	e.lazyConnMgr = nil
-	e.lazyConnMgrMu.Unlock()
 
 	for _, peerID := range e.peerStore.PeersPubKey() {
 		e.peerStore.PeerConnOpen(ctx, peerID)

client/internal/conn_mgr_test.go

@@ -1,21 +1,10 @@
 package internal
 
 import (
-	"context"
-	"net"
-	"net/netip"
 	"os"
-	"sync"
 	"testing"
-	"time"
 
-	"golang.zx2c4.com/wireguard/wgctrl/wgtypes"
-
-	"github.com/netbirdio/netbird/client/iface/wgaddr"
 	"github.com/netbirdio/netbird/client/internal/lazyconn"
-	"github.com/netbirdio/netbird/client/internal/peer"
-	"github.com/netbirdio/netbird/client/internal/peerstore"
-	"github.com/netbirdio/netbird/monotime"
 )
 
 func TestResolveLazyForce(t *testing.T) {
@@ -49,58 +38,3 @@ func TestResolveLazyForce(t *testing.T) {
 		})
 	}
 }
-
-type mockLazyWGIface struct{}
-
-func (mockLazyWGIface) RemovePeer(string) error { return nil }
-func (mockLazyWGIface) UpdatePeer(string, []netip.Prefix, time.Duration, *net.UDPAddr, *wgtypes.Key) error {
-	return nil
-}
-func (mockLazyWGIface) IsUserspaceBind() bool                    { return false }
-func (mockLazyWGIface) Address() wgaddr.Address                  { return wgaddr.Address{} }
-func (mockLazyWGIface) LastActivities() map[string]monotime.Time { return nil }
-func (mockLazyWGIface) MTU() uint16                              { return 1280 }
-
-// TestConnMgr_ActivatePeerConcurrentWithLifecycle exercises ActivatePeer from
-// non-engine goroutines (the DNS warm-up path) racing the manager lifecycle,
-// which stays on the engine loop. Run with -race: it fails if ActivatePeer
-// still requires engine.syncMsgMux for safety.
-func TestConnMgr_ActivatePeerConcurrentWithLifecycle(t *testing.T) {
-	t.Setenv(lazyconn.EnvLazyConn, "on")
-
-	status := peer.NewRecorder("https://mgm")
-	store := peerstore.NewConnStore()
-	connMgr := NewConnMgr(&EngineConfig{}, status, store, mockLazyWGIface{})
-
-	conn := newTestPeerConn(t, "peerA")
-
-	ctx, cancel := context.WithCancel(context.Background())
-	defer cancel()
-	connMgr.Start(ctx)
-
-	done := make(chan struct{})
-	var wg sync.WaitGroup
-	for range 4 {
-		wg.Add(1)
-		go func() {
-			defer wg.Done()
-			for {
-				select {
-				case <-done:
-					return
-				default:
-					connMgr.ActivatePeer(ctx, conn)
-				}
-			}
-		}()
-	}
-
-	// Let the activators spin against the started manager, then tear it down
-	// underneath them and let them spin against the stopped manager.
-	time.Sleep(100 * time.Millisecond)
-	connMgr.Close()
-	time.Sleep(50 * time.Millisecond)
-
-	close(done)
-	wg.Wait()
-}

client/internal/connect.go

@@ -257,10 +257,7 @@ func (c *ConnectClient) run(mobileDependency MobileDependency, runningChan chan
 		log.Errorf("failed to clean up temporary installer file: %v", err)
 	}
 
-	defer func() {
+	defer c.statusRecorder.ClientStop()
-		c.statusRecorder.SetSessionExpiresAt(time.Time{})
-		c.statusRecorder.ClientStop()
-	}()
 	operation := func() error {
 		// if context cancelled we not start new backoff cycle
 		if c.ctx.Err() != nil {
@@ -280,15 +277,6 @@ func (c *ConnectClient) run(mobileDependency MobileDependency, runningChan chan
 		log.Debugf("connecting to the Management service %s", c.config.ManagementURL.Host)
 		mgmClient, err := mgm.NewClient(engineCtx, c.config.ManagementURL.Host, myPrivateKey, mgmTlsEnabled)
 		if err != nil {
-			// On daemon shutdown / Down() the parent context is cancelled
-			// and the dial fails with "context canceled". Wrapping that
-			// into state would leave the snapshot stuck at Connecting+err
-			// until the backoff loop wakes up — instead let the operation
-			// return cleanly so the deferred state.Set(StatusIdle) takes
-			// effect on the next iteration.
-			if c.ctx.Err() != nil {
-				return nil
-			}
 			return wrapErr(gstatus.Errorf(codes.FailedPrecondition, "failed connecting to Management Service : %s", err))
 		}
 		mgmNotifier := statusRecorderToMgmConnStateNotifier(c.statusRecorder)
@@ -427,10 +415,6 @@ func (c *ConnectClient) run(mobileDependency MobileDependency, runningChan chan
 			return wrapErr(err)
 		}
 
-		// Seed the session-expiry deadline from the LoginResponse. Subsequent
-		// changes flow in through SyncResponse and are applied in handleSync.
-		engine.ApplySessionDeadline(loginResp.GetSessionExpiresAt())
-
 		log.Infof("Netbird engine started, the IP is: %s", peerConfig.GetAddress())
 		state.Set(StatusConnected)
 
@@ -467,10 +451,6 @@ func (c *ConnectClient) run(mobileDependency MobileDependency, runningChan chan
 	}
 
 	c.statusRecorder.ClientStart()
-	// Wrap the backoff with c.ctx so Down()/actCancel propagates into the
-	// inter-attempt sleep — otherwise a 15s MaxInterval can keep the retry
-	// loop alive long after the caller asked to give up, leaving the
-	// status stream stuck at Connecting.
 	err = backoff.Retry(operation, backoff.WithContext(backOff, c.ctx))
 	if err != nil {
 		log.Debugf("exiting client retry loop due to unrecoverable error: %s", err)
@@ -621,7 +601,6 @@ func createEngineConfig(key wgtypes.Key, config *profilemanager.Config, peerConf
 		BlockLANAccess:      config.BlockLANAccess,
 		BlockInbound:        config.BlockInbound,
 		DisableIPv6:         config.DisableIPv6,
-		SyncMessageVersion:  config.SyncMessageVersion,
 
 		LazyConnection: lazyconn.ParseState(config.LazyConnection),
 
@@ -697,7 +676,6 @@ func loginToManagement(ctx context.Context, client mgm.Client, pubSSHKey []byte,
 		config.BlockLANAccess,
 		config.BlockInbound,
 		config.DisableIPv6,
-		config.SyncMessageVersion,
 		config.EnableSSHRoot,
 		config.EnableSSHSFTP,
 		config.EnableSSHLocalPortForwarding,

client/internal/debug/debug.go

@@ -229,16 +229,9 @@ scutil_dns.txt (macOS only):
 
 const (
 	clientLogFile = "client.log"
-	uiLogFile     = "gui-client.log"
 	errorLogFile  = "netbird.err"
 	stdoutLogFile = "netbird.out"
 
-	// Rotated-log glob prefixes (base log name without extension) passed to
-	// addRotatedLogFiles. The daemon's own log and the GUI log live in the same
-	// dir, so the prefixes must be disjoint to keep their rotated siblings apart.
-	clientLogPrefix = "client"
-	uiLogPrefix     = "gui-client"
-
 	darwinErrorLogPath  = "/var/log/netbird.out.log"
 	darwinStdoutLogPath = "/var/log/netbird.err.log"
 )
@@ -256,7 +249,6 @@ type BundleGenerator struct {
 	statusRecorder *peer.Status
 	syncResponse   *mgmProto.SyncResponse
 	logPath        string
-	uiLogPath      string
 	tempDir        string
 	statePath      string
 	cpuProfile     []byte
@@ -284,7 +276,6 @@ type GeneratorDependencies struct {
 	StatusRecorder *peer.Status
 	SyncResponse   *mgmProto.SyncResponse
 	LogPath        string
-	UILogPath      string // Absolute path to the desktop UI's gui-client.log, reported via RegisterUILog. Empty if no UI registered one.
 	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
@@ -309,7 +300,6 @@ func NewBundleGenerator(deps GeneratorDependencies, cfg BundleConfig) *BundleGen
 		statusRecorder: deps.StatusRecorder,
 		syncResponse:   deps.SyncResponse,
 		logPath:        deps.LogPath,
-		uiLogPath:      deps.UILogPath,
 		tempDir:        deps.TempDir,
 		statePath:      deps.StatePath,
 		cpuProfile:     deps.CPUProfile,
@@ -421,10 +411,6 @@ func (g *BundleGenerator) createArchive() error {
 		log.Errorf("failed to add logs to debug bundle: %v", err)
 	}
 
-	if err := g.addUILog(); err != nil {
-		log.Errorf("failed to add UI log to debug bundle: %v", err)
-	}
-
 	if err := g.addUpdateLogs(); err != nil {
 		log.Errorf("failed to add updater logs: %v", err)
 	}
@@ -480,6 +466,7 @@ func (g *BundleGenerator) addStatus() error {
 
 		fullStatus := g.statusRecorder.GetFullStatus()
 		protoFullStatus := nbstatus.ToProtoFullStatus(fullStatus)
+		protoFullStatus.Events = g.statusRecorder.GetEventHistory()
 		overview := nbstatus.ConvertToStatusOutputOverview(protoFullStatus, nbstatus.ConvertOptions{
 			Anonymize:     g.anonymize,
 			ProfileName:   profName,
@@ -676,7 +663,6 @@ func (g *BundleGenerator) addCommonConfigFields(configContent *strings.Builder)
 	configContent.WriteString(fmt.Sprintf("BlockLANAccess: %v\n", g.internalConfig.BlockLANAccess))
 	configContent.WriteString(fmt.Sprintf("BlockInbound: %v\n", g.internalConfig.BlockInbound))
 	configContent.WriteString(fmt.Sprintf("DisableIPv6: %v\n", g.internalConfig.DisableIPv6))
-	configContent.WriteString(fmt.Sprintf("SyncMessageVersion: %v\n", g.internalConfig.SyncMessageVersion))
 
 	if g.internalConfig.DisableNotifications != nil {
 		configContent.WriteString(fmt.Sprintf("DisableNotifications: %v\n", *g.internalConfig.DisableNotifications))
@@ -1000,7 +986,7 @@ func (g *BundleGenerator) addLogfile() error {
 		return fmt.Errorf("add client log file to zip: %w", err)
 	}
 
-	g.addRotatedLogFiles(logDir, clientLogPrefix)
+	g.addRotatedLogFiles(logDir)
 
 	stdErrLogPath := filepath.Join(logDir, errorLogFile)
 	stdoutLogPath := filepath.Join(logDir, stdoutLogFile)
@@ -1020,25 +1006,6 @@ func (g *BundleGenerator) addLogfile() error {
 	return nil
 }
 
-// addUILog adds the desktop UI's gui-client.log (and its rotated siblings) to
-// the bundle. The path is reported by the UI via RegisterUILog; empty when no
-// UI registered one (e.g. headless / server). Missing file is non-fatal — the
-// UI only writes it while the daemon is in debug, so it's often absent.
-func (g *BundleGenerator) addUILog() error {
-	if g.uiLogPath == "" {
-		log.Debugf("no UI log path registered, skipping in debug bundle")
-		return nil
-	}
-
-	if err := g.addSingleLogfile(g.uiLogPath, uiLogFile); err != nil {
-		return fmt.Errorf("add UI log file to zip: %w", err)
-	}
-
-	g.addRotatedLogFiles(filepath.Dir(g.uiLogPath), uiLogPrefix)
-
-	return nil
-}
-
 // addSingleLogfile adds a single log file to the archive
 func (g *BundleGenerator) addSingleLogfile(logPath, targetName string) error {
 	logFile, err := os.Open(logPath)
@@ -1111,16 +1078,14 @@ func (g *BundleGenerator) addSingleLogFileGz(logPath, targetName string) error {
 	return nil
 }
 
-// addRotatedLogFiles adds rotated log files to the bundle based on logFileCount.
+// addRotatedLogFiles adds rotated log files to the bundle based on logFileCount
-// prefix is the base log name without extension (e.g. "client", "gui-client");
+func (g *BundleGenerator) addRotatedLogFiles(logDir string) {
-// the glob matches both files rotated by us and by logrotate on linux.
-func (g *BundleGenerator) addRotatedLogFiles(logDir, prefix string) {
 	if g.logFileCount == 0 {
 		return
 	}
 
-	// This pattern matches both logs rotated by us and logrotate on linux
+	// This regex will match both logs rotated by us and logrotate on linux
-	pattern := filepath.Join(logDir, prefix+"*.log.*")
+	pattern := filepath.Join(logDir, "client*.log.*")
 	files, err := filepath.Glob(pattern)
 	if err != nil {
 		log.Warnf("failed to glob rotated logs: %v", err)

client/internal/debug/debug_logfiles_test.go

@@ -40,25 +40,6 @@ func TestAddRotatedLogFiles_PicksUpAllVariants(t *testing.T) {
 	require.NotContains(t, names, "other.log", "unrelated files should not be in bundle")
 }
 
-// TestAddRotatedLogFiles_GUIPrefix asserts the prefix parameter scopes the glob
-// to the GUI log: gui-client.log.* rotated siblings are picked up and the
-// daemon's own client.log.* are not (and vice versa, covered above). This is
-// the load-bearing check for the gui-client.log bundle collection — the old
-// "client*.log.*" glob would have missed gui-client rotations.
-func TestAddRotatedLogFiles_GUIPrefix(t *testing.T) {
-	dir := t.TempDir()
-
-	writeFile(t, filepath.Join(dir, "gui-client.log.1"), "gui rotated\n")
-	writeGzFile(t, filepath.Join(dir, "gui-client.log.2.gz"), "gui rotated gz\n")
-	writeFile(t, filepath.Join(dir, "client.log.1"), "daemon rotated\n")
-
-	names := runAddRotatedLogFilesPrefix(t, dir, "gui-client", 10)
-
-	require.Contains(t, names, "gui-client.log.1", "gui-client rotated file should be in bundle")
-	require.Contains(t, names, "gui-client.log.2.gz", "gui-client gz rotated file should be in bundle")
-	require.NotContains(t, names, "client.log.1", "daemon rotated file must not match the gui-client prefix")
-}
-
 // TestAddRotatedLogFiles_RespectsLogFileCount asserts that only the newest
 // logFileCount rotated files are bundled, ordered by mtime.
 func TestAddRotatedLogFiles_RespectsLogFileCount(t *testing.T) {
@@ -86,10 +67,6 @@ func TestAddRotatedLogFiles_RespectsLogFileCount(t *testing.T) {
 // runAddRotatedLogFiles calls addRotatedLogFiles against a fresh in-memory
 // zip writer and returns the set of entry names that ended up in the archive.
 func runAddRotatedLogFiles(t *testing.T, dir string, logFileCount uint32) map[string]struct{} {
-	return runAddRotatedLogFilesPrefix(t, dir, "client", logFileCount)
-}
-
-func runAddRotatedLogFilesPrefix(t *testing.T, dir, prefix string, logFileCount uint32) map[string]struct{} {
 	t.Helper()
 
 	var buf bytes.Buffer
@@ -97,7 +74,7 @@ func runAddRotatedLogFilesPrefix(t *testing.T, dir, prefix string, logFileCount
 		archive:      zip.NewWriter(&buf),
 		logFileCount: logFileCount,
 	}
-	g.addRotatedLogFiles(dir, prefix)
+	g.addRotatedLogFiles(dir)
 	require.NoError(t, g.archive.Close())
 
 	zr, err := zip.NewReader(bytes.NewReader(buf.Bytes()), int64(buf.Len()))

client/internal/debug/debug_test.go

@@ -887,8 +887,6 @@ func TestAddConfig_AllFieldsCovered(t *testing.T) {
 		ClientCertKeyPath:             "/tmp/key",
 		LazyConnection:                "on",
 		MTU:                           1280,
-		DisableIPv6:                   true,
-		SyncMessageVersion:            func(v int) *int { return &v }(1),
 	}
 
 	for _, anonymize := range []bool{false, true} {

client/internal/dns/local/local.go

@@ -6,7 +6,6 @@ import (
 	"fmt"
 	"net"
 	"net/netip"
-	"os"
 	"slices"
 	"strings"
 	"sync"
@@ -37,43 +36,7 @@ type resolver interface {
 // record is left alone (it points at something outside our mesh, e.g.
 // a non-peer upstream).
 type PeerConnectivity interface {
-	IsConnectedByIP(ip netip.Addr) (known, connected bool)
+	IsConnectedByIP(ip string) (known, connected bool)
-}
-
-// PeerActivator wakes lazy-connection peers on demand. The local resolver calls
-// it with the tunnel IPs an answer points at, so a peer that is idle (lazily
-// disconnected) starts connecting at DNS-resolution time rather than racing the
-// client's first request packet. nil disables warm-up.
-type PeerActivator interface {
-	// ActivatePeersByIP triggers wake-up for the peer(s) owning addrs and blocks
-	// until one is connected or ctx (a short per-query budget) expires. It is a
-	// fast no-op for unknown or already-connected addresses.
-	ActivatePeersByIP(ctx context.Context, addrs []netip.Addr)
-}
-
-const (
-	defaultLazyWarmupTimeout = 2 * time.Second
-	envLazyWarmupTimeout     = "NB_DNS_LAZY_WARMUP_TIMEOUT"
-)
-
-// lazyWarmupTimeoutFromEnv returns the per-query budget for waking a
-// lazy-connection peer a DNS answer points at. Tunable via
-// NB_DNS_LAZY_WARMUP_TIMEOUT (a Go duration). Parsed once at construction time.
-func lazyWarmupTimeoutFromEnv() time.Duration {
-	v := os.Getenv(envLazyWarmupTimeout)
-	if v == "" {
-		return defaultLazyWarmupTimeout
-	}
-	d, err := time.ParseDuration(v)
-	if err != nil {
-		log.Warnf("invalid %s value %q, using default %s: %v", envLazyWarmupTimeout, v, defaultLazyWarmupTimeout, err)
-		return defaultLazyWarmupTimeout
-	}
-	if d <= 0 {
-		log.Warnf("non-positive %s value %q, using default %s", envLazyWarmupTimeout, v, defaultLazyWarmupTimeout)
-		return defaultLazyWarmupTimeout
-	}
-	return d
 }
 
 type Resolver struct {
@@ -88,12 +51,6 @@ type Resolver struct {
 	// filter and preserves the legacy "return whatever is registered"
 	// behaviour for callers that never wire a status source.
 	peerConn PeerConnectivity
-	// peerActivator, when non-nil, is called at resolution time to warm the
-	// lazy connection to the peer(s) an answer points at. nil disables warm-up.
-	peerActivator PeerActivator
-	// warmupTimeout is the per-query budget for the lazy-connection warm-up
-	// wait, resolved from the environment once at construction time.
-	warmupTimeout time.Duration
 
 	ctx    context.Context
 	cancel context.CancelFunc
@@ -102,12 +59,11 @@ type Resolver struct {
 func NewResolver() *Resolver {
 	ctx, cancel := context.WithCancel(context.Background())
 	return &Resolver{
-		records:       make(map[dns.Question][]dns.RR),
+		records: make(map[dns.Question][]dns.RR),
-		domains:       make(map[domain.Domain]struct{}),
+		domains: make(map[domain.Domain]struct{}),
-		zones:         make(map[domain.Domain]bool),
+		zones:   make(map[domain.Domain]bool),
-		warmupTimeout: lazyWarmupTimeoutFromEnv(),
+		ctx:     ctx,
-		ctx:           ctx,
+		cancel:  cancel,
-		cancel:        cancel,
 	}
 }
 
@@ -120,14 +76,6 @@ func (d *Resolver) SetPeerConnectivity(p PeerConnectivity) {
 	d.peerConn = p
 }
 
-// SetPeerActivator wires the DNS-time lazy-connection warm-up. Pass nil to
-// disable. Safe to call multiple times; the latest value wins.
-func (d *Resolver) SetPeerActivator(a PeerActivator) {
-	d.mu.Lock()
-	defer d.mu.Unlock()
-	d.peerActivator = a
-}
-
 func (d *Resolver) MatchSubdomains() bool {
 	return true
 }
@@ -174,9 +122,6 @@ func (d *Resolver) ServeDNS(w dns.ResponseWriter, r *dns.Msg) {
 	replyMessage.RecursionAvailable = true
 
 	result := d.lookupRecords(logger, question)
-	// Warm before filtering: activation flips a lazily-idle target to connected,
-	// which then lets it survive the disconnected-peer filter below.
-	d.warmLazyPeers(question, result.records)
 	result.records = d.filterDisconnectedPeerAnswers(logger, question, result.records)
 	replyMessage.Authoritative = !result.hasExternalData
 	replyMessage.Answer = result.records
@@ -550,8 +495,8 @@ func (d *Resolver) filterDisconnectedPeerAnswers(logger *log.Entry, question dns
 	kept := make([]dns.RR, 0, len(records))
 	var dropped int
 	for _, rr := range records {
-		ip, ok := extractRecordAddr(rr)
+		ip := extractRecordIP(rr)
-		if !ok {
+		if ip == "" {
 			kept = append(kept, rr)
 			continue
 		}
@@ -573,57 +518,22 @@ func (d *Resolver) filterDisconnectedPeerAnswers(logger *log.Entry, question dns
 	return kept
 }
 
-// warmLazyPeers triggers lazy-connection wake-up for the peers a resolved
+// extractRecordIP returns the dotted-decimal / colon-hex IP carried by
-// answer points at and waits briefly for one to connect, so the caller's first
+// an A or AAAA record, or "" for any other record type.
-// request doesn't race the connection establishment. Warm-up is scoped to
+func extractRecordIP(rr dns.RR) string {
-// match-only (non-authoritative) zones — the synthesized private-service zones
-// and user-created zones whose records point at specific peers. The account's
-// peer zone is authoritative, so plain peer-name lookups never trigger warm-up;
-// otherwise resolving any peer's name would wake its idle connection, defeating
-// laziness mesh-wide. No-op when no activator is wired (lazy connections
-// disabled) or the answer carries no peer IPs.
-func (d *Resolver) warmLazyPeers(question dns.Question, records []dns.RR) {
-	if len(records) < 2 {
-		return
-	}
-	d.mu.RLock()
-	activator := d.peerActivator
-	var nonAuth, found bool
-	if activator != nil {
-		nonAuth, found = d.findZone(question.Name)
-	}
-	d.mu.RUnlock()
-	if activator == nil || !found || !nonAuth {
-		return
-	}
-
-	var addrs []netip.Addr
-	for _, rr := range records {
-		if addr, ok := extractRecordAddr(rr); ok {
-			addrs = append(addrs, addr)
-		}
-	}
-	if len(addrs) == 0 {
-		return
-	}
-
-	ctx, cancel := context.WithTimeout(d.ctx, d.warmupTimeout)
-	defer cancel()
-	activator.ActivatePeersByIP(ctx, addrs)
-}
-
-// extractRecordAddr returns the IP address carried by an A or AAAA record.
-// ok is false for any other record type or a record with no address.
-func extractRecordAddr(rr dns.RR) (netip.Addr, bool) {
 	switch r := rr.(type) {
 	case *dns.A:
-		addr, ok := netip.AddrFromSlice(r.A)
+		if r.A == nil {
-		return addr.Unmap(), ok
+			return ""
+		}
+		return r.A.String()
 	case *dns.AAAA:
-		addr, ok := netip.AddrFromSlice(r.AAAA)
+		if r.AAAA == nil {
-		return addr.Unmap(), ok
+			return ""
+		}
+		return r.AAAA.String()
 	}
-	return netip.Addr{}, false
+	return ""
 }
 
 // Update replaces all zones and their records

client/internal/dns/local/local_test.go

@@ -37,8 +37,8 @@ type mockPeerConnectivity struct {
 	byIP map[string]struct{ known, connected bool }
 }
 
-func (m mockPeerConnectivity) IsConnectedByIP(ip netip.Addr) (known, connected bool) {
+func (m mockPeerConnectivity) IsConnectedByIP(ip string) (known, connected bool) {
-	v, ok := m.byIP[ip.String()]
+	v, ok := m.byIP[ip]
 	if !ok {
 		return false, false
 	}

client/internal/dns/local/warmup_test.go

@@ -1,204 +0,0 @@
-package local
-
-import (
-	"context"
-	"net"
-	"net/netip"
-	"sync"
-	"testing"
-	"time"
-
-	"github.com/miekg/dns"
-	"github.com/stretchr/testify/assert"
-	"github.com/stretchr/testify/require"
-
-	"github.com/netbirdio/netbird/client/internal/dns/test"
-	nbdns "github.com/netbirdio/netbird/dns"
-)
-
-// recordingActivator records the addresses it was asked to warm and returns
-// immediately, so ServeDNS is not blocked by the test.
-type recordingActivator struct {
-	mu     sync.Mutex
-	called bool
-	addrs  []netip.Addr
-}
-
-func (r *recordingActivator) ActivatePeersByIP(_ context.Context, addrs []netip.Addr) {
-	r.mu.Lock()
-	defer r.mu.Unlock()
-	r.called = true
-	r.addrs = append(r.addrs, addrs...)
-}
-
-func serveA(t *testing.T, resolver *Resolver, name string) *dns.Msg {
-	t.Helper()
-	var resp *dns.Msg
-	w := &test.MockResponseWriter{WriteMsgFunc: func(m *dns.Msg) error { resp = m; return nil }}
-	resolver.ServeDNS(w, new(dns.Msg).SetQuestion(name, dns.TypeA))
-	return resp
-}
-
-// serviceZone registers rec in a match-only (non-authoritative) zone, the shape
-// the synthesized private-service zones arrive in.
-func serviceZone(t *testing.T, resolver *Resolver, zone string, records ...nbdns.SimpleRecord) {
-	t.Helper()
-	resolver.Update([]nbdns.CustomZone{{
-		Domain:           zone,
-		Records:          records,
-		NonAuthoritative: true,
-	}})
-}
-
-func TestLocalResolver_WarmsLazyPeerOnResolve(t *testing.T) {
-	// Warm-up fires only for multi-record answers (the HA / round-robin shape of
-	// the synthesized private-service zones), so register two peer targets.
-	const name = "svc.proxy.netbird.cloud."
-	recs := []nbdns.SimpleRecord{
-		{Name: name, Type: 1, Class: nbdns.DefaultClass, TTL: 300, RData: "100.64.0.7"},
-		{Name: name, Type: 1, Class: nbdns.DefaultClass, TTL: 300, RData: "100.64.0.8"},
-	}
-	resolver := NewResolver()
-	serviceZone(t, resolver, "proxy.netbird.cloud", recs...)
-
-	act := &recordingActivator{}
-	resolver.SetPeerActivator(act)
-
-	resp := serveA(t, resolver, name)
-	require.NotNil(t, resp, "resolver must answer")
-	require.NotEmpty(t, resp.Answer, "answer must carry the A records")
-
-	act.mu.Lock()
-	defer act.mu.Unlock()
-	assert.True(t, act.called, "activator must be invoked for a multi-record service-zone answer")
-	assert.Contains(t, act.addrs, netip.MustParseAddr("100.64.0.7"), "activator must receive the first peer IP")
-	assert.Contains(t, act.addrs, netip.MustParseAddr("100.64.0.8"), "activator must receive the second peer IP")
-}
-
-func TestLocalResolver_NoWarmupForSingleRecord(t *testing.T) {
-	// A single-record answer does not trigger warm-up; the resolver only warms
-	// multi-record answers.
-	rec := nbdns.SimpleRecord{Name: "svc.proxy.netbird.cloud.", Type: 1, Class: nbdns.DefaultClass, TTL: 300, RData: "100.64.0.7"}
-	resolver := NewResolver()
-	serviceZone(t, resolver, "proxy.netbird.cloud", rec)
-
-	act := &recordingActivator{}
-	resolver.SetPeerActivator(act)
-
-	resp := serveA(t, resolver, rec.Name)
-	require.NotNil(t, resp, "resolver must answer")
-	require.NotEmpty(t, resp.Answer, "answer must carry the A record")
-
-	act.mu.Lock()
-	defer act.mu.Unlock()
-	assert.False(t, act.called, "activator must not be invoked for a single-record answer")
-}
-
-func TestLocalResolver_NoActivatorNoWarmup(t *testing.T) {
-	// With no activator wired the resolver behaves exactly as before.
-	rec := nbdns.SimpleRecord{Name: "svc.proxy.netbird.cloud.", Type: 1, Class: nbdns.DefaultClass, TTL: 300, RData: "100.64.0.7"}
-	resolver := NewResolver()
-	serviceZone(t, resolver, "proxy.netbird.cloud", rec)
-
-	resp := serveA(t, resolver, rec.Name)
-	require.NotNil(t, resp, "resolver must still answer without an activator")
-	require.NotEmpty(t, resp.Answer, "answer must carry the A record")
-}
-
-func TestLocalResolver_NoWarmupForMissingRecord(t *testing.T) {
-	// A query that resolves to nothing must not invoke the activator (no IPs).
-	resolver := NewResolver()
-	serviceZone(t, resolver, "proxy.netbird.cloud",
-		nbdns.SimpleRecord{Name: "svc.proxy.netbird.cloud.", Type: 1, Class: nbdns.DefaultClass, TTL: 300, RData: "100.64.0.7"})
-
-	act := &recordingActivator{}
-	resolver.SetPeerActivator(act)
-
-	serveA(t, resolver, "absent.proxy.netbird.cloud.")
-
-	act.mu.Lock()
-	defer act.mu.Unlock()
-	assert.False(t, act.called, "activator must not be invoked when there is no answer")
-}
-
-func TestLocalResolver_NoWarmupInAuthoritativeZone(t *testing.T) {
-	// The account's peer zone is authoritative; resolving a peer's name there
-	// must not wake its lazy connection — warm-up is scoped to match-only
-	// (non-authoritative) zones such as the synthesized private-service zones.
-	// Use a multi-record answer so the authoritative-zone scoping is the only
-	// reason warm-up is skipped, not the single-record guard.
-	const name = "peer.netbird.cloud."
-	recs := []nbdns.SimpleRecord{
-		{Name: name, Type: 1, Class: nbdns.DefaultClass, TTL: 300, RData: "100.64.0.9"},
-		{Name: name, Type: 1, Class: nbdns.DefaultClass, TTL: 300, RData: "100.64.0.10"},
-	}
-	resolver := NewResolver()
-	resolver.Update([]nbdns.CustomZone{{
-		Domain:  "netbird.cloud",
-		Records: recs,
-	}})
-
-	act := &recordingActivator{}
-	resolver.SetPeerActivator(act)
-
-	resp := serveA(t, resolver, name)
-	require.NotNil(t, resp, "resolver must answer")
-	require.NotEmpty(t, resp.Answer, "answer must carry the A records")
-
-	act.mu.Lock()
-	defer act.mu.Unlock()
-	assert.False(t, act.called, "activator must not be invoked for authoritative-zone answers")
-}
-
-func TestLazyWarmupTimeoutFromEnv(t *testing.T) {
-	tests := []struct {
-		name   string
-		value  string
-		envSet bool
-		want   time.Duration
-	}{
-		{name: "unset uses default", want: defaultLazyWarmupTimeout},
-		{name: "valid overrides", value: "5s", envSet: true, want: 5 * time.Second},
-		{name: "invalid falls back", value: "not-a-duration", envSet: true, want: defaultLazyWarmupTimeout},
-		{name: "negative falls back", value: "-1s", envSet: true, want: defaultLazyWarmupTimeout},
-		{name: "zero falls back", value: "0s", envSet: true, want: defaultLazyWarmupTimeout},
-	}
-
-	for _, tt := range tests {
-		t.Run(tt.name, func(t *testing.T) {
-			if tt.envSet {
-				t.Setenv(envLazyWarmupTimeout, tt.value)
-			}
-			assert.Equal(t, tt.want, lazyWarmupTimeoutFromEnv())
-			assert.Equal(t, tt.want, NewResolver().warmupTimeout, "constructor must resolve the timeout once")
-		})
-	}
-}
-
-func TestExtractRecordAddr(t *testing.T) {
-	t.Run("A record yields unmapped v4", func(t *testing.T) {
-		// net.ParseIP returns the 16-byte v4-in-v6 form, the same shape
-		// miekg/dns stores after parsing an A record; the extracted address
-		// must compare equal to a plain v4 netip.Addr.
-		addr, ok := extractRecordAddr(&dns.A{A: net.ParseIP("100.64.0.7")})
-		require.True(t, ok)
-		assert.True(t, addr.Is4())
-		assert.Equal(t, netip.MustParseAddr("100.64.0.7"), addr)
-	})
-
-	t.Run("AAAA record yields v6", func(t *testing.T) {
-		addr, ok := extractRecordAddr(&dns.AAAA{AAAA: net.ParseIP("fd00::1")})
-		require.True(t, ok)
-		assert.Equal(t, netip.MustParseAddr("fd00::1"), addr)
-	})
-
-	t.Run("A record without address", func(t *testing.T) {
-		_, ok := extractRecordAddr(&dns.A{})
-		assert.False(t, ok)
-	})
-
-	t.Run("non-address record", func(t *testing.T) {
-		_, ok := extractRecordAddr(&dns.CNAME{Target: "target.netbird.cloud."})
-		assert.False(t, ok)
-	})
-}

client/internal/dns/mock_server.go

@@ -8,7 +8,6 @@ import (
 	"github.com/miekg/dns"
 
 	dnsconfig "github.com/netbirdio/netbird/client/internal/dns/config"
-	"github.com/netbirdio/netbird/client/internal/dns/local"
 	nbdns "github.com/netbirdio/netbird/dns"
 	"github.com/netbirdio/netbird/route"
 	"github.com/netbirdio/netbird/shared/management/domain"
@@ -93,11 +92,6 @@ func (m *MockServer) SetFirewall(Firewall) {
 	// Mock implementation - no-op
 }
 
-// SetPeerActivator mock implementation of SetPeerActivator from Server interface
-func (m *MockServer) SetPeerActivator(local.PeerActivator) {
-	// Mock implementation - no-op
-}
-
 // BeginBatch mock implementation of BeginBatch from Server interface
 func (m *MockServer) BeginBatch() {
 	// Mock implementation - no-op

client/internal/dns/server.go

@@ -82,7 +82,6 @@ type Server interface {
 	PopulateManagementDomain(mgmtURL *url.URL) error
 	SetRouteSources(selected, active func() route.HAMap)
 	SetFirewall(Firewall)
-	SetPeerActivator(local.PeerActivator)
 }
 
 type nsGroupsByDomain struct {
@@ -492,13 +491,6 @@ func (s *DefaultServer) SetFirewall(fw Firewall) {
 	}
 }
 
-// SetPeerActivator wires the DNS-time lazy-connection warm-up on the local
-// resolver. Injected after the connection manager exists (it does not at
-// DNS-server construction time). Pass nil to disable.
-func (s *DefaultServer) SetPeerActivator(a local.PeerActivator) {
-	s.localResolver.SetPeerActivator(a)
-}
-
 // Stop stops the server
 func (s *DefaultServer) Stop() {
 	s.ctxCancel()
@@ -1443,11 +1435,11 @@ type localPeerConnectivity struct {
 
 // IsConnectedByIP looks the IP up in the peerstore and surfaces both
 // the known and connected bits. Used by Resolver.filterDisconnectedPeerAnswers.
-func (l localPeerConnectivity) IsConnectedByIP(ip netip.Addr) (known, connected bool) {
+func (l localPeerConnectivity) IsConnectedByIP(ip string) (known, connected bool) {
 	if l.status == nil {
 		return false, false
 	}
-	state, ok := l.status.PeerStateByIP(ip.String())
+	state, ok := l.status.PeerStateByIP(ip)
 	if !ok {
 		return false, false
 	}

client/internal/dns/service_listener.go

@@ -292,16 +292,18 @@ func (s *serviceViaListener) generateFreePort() (uint16, error) {
 		return customPort, nil
 	}
 
-	probeListener, err := net.ListenUDP("udp4", &net.UDPAddr{})
+	udpAddr := net.UDPAddrFromAddrPort(netip.MustParseAddrPort("0.0.0.0:0"))
+	probeListener, err := net.ListenUDP("udp", udpAddr)
 	if err != nil {
 		log.Debugf("failed to bind random port for DNS: %s", err)
 		return 0, err
 	}
 
-	port := uint16(probeListener.LocalAddr().(*net.UDPAddr).Port)
+	addrPort := netip.MustParseAddrPort(probeListener.LocalAddr().String()) // might panic if address is incorrect
-	if err = probeListener.Close(); err != nil {
+	err = probeListener.Close()
+	if err != nil {
 		log.Debugf("failed to free up DNS port: %s", err)
 		return 0, err
 	}
-	return port, nil
+	return addrPort.Port(), nil
 }

client/internal/dns_peer_activator.go

@@ -1,76 +0,0 @@
-package internal
-
-import (
-	"context"
-	"net/netip"
-	"time"
-
-	"github.com/netbirdio/netbird/client/internal/peer"
-	"github.com/netbirdio/netbird/client/internal/peerstore"
-)
-
-const dnsActivationPollInterval = 50 * time.Millisecond
-
-// dnsPeerActivator wakes lazy-connection peers from the DNS resolution path. It
-// implements dns/local.PeerActivator. DNS queries run on their own goroutines,
-// so it only touches state that is safe for concurrent use — ConnMgr.ActivatePeer,
-// peerstore.Store and peer.Status — and never takes the engine's syncMsgMux,
-// keeping DNS resolution from contending with network-map processing.
-type dnsPeerActivator struct {
-	connMgr   *ConnMgr
-	peerStore *peerstore.Store
-	status    *peer.Status
-	// ctx is the engine's long-lived context. The connection dial is tied to it
-	// (not the per-query DNS wait budget) so a handshake that outlasts the wait
-	// still completes in the background rather than being cancelled at the deadline.
-	ctx context.Context
-}
-
-// ActivatePeersByIP triggers wake-up for the peer(s) owning addrs and waits
-// until one is connected or ctx (the per-query DNS wait budget) expires.
-// Activation itself is tied to the engine's long-lived context so the dial
-// survives a wait that times out. Unknown or already-connected addresses are
-// skipped, so the steady-state (warm) path adds no latency.
-func (a *dnsPeerActivator) ActivatePeersByIP(ctx context.Context, addrs []netip.Addr) {
-	if a == nil || a.connMgr == nil {
-		return
-	}
-
-	var pending []string
-	for _, addr := range addrs {
-		ip := addr.String()
-		st, ok := a.status.PeerStateByIP(ip)
-		if !ok || st.ConnStatus == peer.StatusConnected {
-			continue
-		}
-		conn, ok := a.peerStore.PeerConn(st.PubKey)
-		if !ok {
-			continue
-		}
-		a.connMgr.ActivatePeer(a.ctx, conn)
-		pending = append(pending, ip)
-	}
-
-	if len(pending) == 0 {
-		return
-	}
-	a.waitConnected(ctx, pending)
-}
-
-// waitConnected blocks until any of ips reports a connected peer or ctx expires.
-func (a *dnsPeerActivator) waitConnected(ctx context.Context, ips []string) {
-	ticker := time.NewTicker(dnsActivationPollInterval)
-	defer ticker.Stop()
-	for {
-		for _, ip := range ips {
-			if st, ok := a.status.PeerStateByIP(ip); ok && st.ConnStatus == peer.StatusConnected {
-				return
-			}
-		}
-		select {
-		case <-ctx.Done():
-			return
-		case <-ticker.C:
-		}
-	}
-}

client/internal/dns_peer_activator_test.go

@@ -1,129 +0,0 @@
-package internal
-
-import (
-	"context"
-	"net/netip"
-	"testing"
-	"time"
-
-	"github.com/stretchr/testify/require"
-
-	"github.com/netbirdio/netbird/client/internal/peer"
-	"github.com/netbirdio/netbird/client/internal/peerstore"
-)
-
-func newTestPeerConn(t *testing.T, key string) *peer.Conn {
-	t.Helper()
-	conn, err := peer.NewConn(peer.ConnConfig{
-		Key:      key,
-		LocalKey: "local",
-		WgConfig: peer.WgConfig{
-			AllowedIps: []netip.Prefix{netip.MustParsePrefix("100.64.0.1/32")},
-		},
-	}, peer.ServiceDependencies{})
-	require.NoError(t, err)
-	return conn
-}
-
-func newTestDNSPeerActivator(t *testing.T) (*dnsPeerActivator, *peer.Status, *peerstore.Store) {
-	t.Helper()
-	status := peer.NewRecorder("https://mgm")
-	store := peerstore.NewConnStore()
-	// ConnMgr without Start: the lazy manager is nil, so ActivatePeer is a
-	// no-op — these tests exercise the activator's skip/wait logic.
-	connMgr := NewConnMgr(&EngineConfig{}, status, store, nil)
-	return &dnsPeerActivator{
-		connMgr:   connMgr,
-		peerStore: store,
-		status:    status,
-		ctx:       context.Background(),
-	}, status, store
-}
-
-func TestDNSPeerActivator_NilSafe(t *testing.T) {
-	var a *dnsPeerActivator
-	a.ActivatePeersByIP(context.Background(), []netip.Addr{netip.MustParseAddr("100.64.0.1")})
-}
-
-// TestDNSPeerActivator_SkipsUnknownAndConnectedPeers verifies the steady-state
-// (warm) path adds no latency: already-connected and unknown addresses never
-// enter the wait loop.
-func TestDNSPeerActivator_SkipsUnknownAndConnectedPeers(t *testing.T) {
-	a, status, store := newTestDNSPeerActivator(t)
-
-	require.NoError(t, status.AddPeer("peerA", "a.netbird.cloud", "100.64.0.1", "fd00::1"))
-	require.NoError(t, status.UpdatePeerState(peer.State{PubKey: "peerA", ConnStatus: peer.StatusConnected}))
-	store.AddPeerConn("peerA", newTestPeerConn(t, "peerA"))
-
-	ctx, cancel := context.WithTimeout(context.Background(), 5*time.Second)
-	defer cancel()
-
-	start := time.Now()
-	a.ActivatePeersByIP(ctx, []netip.Addr{
-		netip.MustParseAddr("100.64.0.1"),  // known, connected -> skipped
-		netip.MustParseAddr("fd00::1"),     // known via IPv6, connected -> skipped
-		netip.MustParseAddr("100.64.0.99"), // unknown -> skipped
-	})
-	require.Less(t, time.Since(start), time.Second, "no pending peer must mean no wait")
-}
-
-// TestDNSPeerActivator_WaitsForPendingPeerToConnect verifies the wait loop
-// returns as soon as a pending peer reports connected, well before the
-// per-query budget expires.
-func TestDNSPeerActivator_WaitsForPendingPeerToConnect(t *testing.T) {
-	a, status, store := newTestDNSPeerActivator(t)
-
-	require.NoError(t, status.AddPeer("peerA", "a.netbird.cloud", "100.64.0.1", ""))
-	store.AddPeerConn("peerA", newTestPeerConn(t, "peerA"))
-
-	go func() {
-		time.Sleep(150 * time.Millisecond)
-		_ = status.UpdatePeerState(peer.State{PubKey: "peerA", ConnStatus: peer.StatusConnected})
-	}()
-
-	ctx, cancel := context.WithTimeout(context.Background(), 10*time.Second)
-	defer cancel()
-
-	start := time.Now()
-	a.ActivatePeersByIP(ctx, []netip.Addr{netip.MustParseAddr("100.64.0.1")})
-	elapsed := time.Since(start)
-
-	require.GreaterOrEqual(t, elapsed, 100*time.Millisecond, "must wait for the pending peer")
-	require.Less(t, elapsed, 5*time.Second, "must return on connect, not at the deadline")
-}
-
-// TestDNSPeerActivator_ReturnsAtBudgetWhenPeerStaysIdle verifies a peer that
-// never connects releases the DNS response at the per-query budget instead of
-// blocking it indefinitely.
-func TestDNSPeerActivator_ReturnsAtBudgetWhenPeerStaysIdle(t *testing.T) {
-	a, status, store := newTestDNSPeerActivator(t)
-
-	require.NoError(t, status.AddPeer("peerA", "a.netbird.cloud", "100.64.0.1", ""))
-	store.AddPeerConn("peerA", newTestPeerConn(t, "peerA"))
-
-	ctx, cancel := context.WithTimeout(context.Background(), 300*time.Millisecond)
-	defer cancel()
-
-	start := time.Now()
-	a.ActivatePeersByIP(ctx, []netip.Addr{netip.MustParseAddr("100.64.0.1")})
-	elapsed := time.Since(start)
-
-	require.GreaterOrEqual(t, elapsed, 250*time.Millisecond, "must wait out the budget for a pending peer")
-	require.Less(t, elapsed, 5*time.Second, "must not block past the budget")
-}
-
-// TestDNSPeerActivator_NoWaitWithoutPeerConn verifies a known-but-idle peer
-// with no connection object in the store is not waited on: there is nothing to
-// activate, so waiting could only ever time out.
-func TestDNSPeerActivator_NoWaitWithoutPeerConn(t *testing.T) {
-	a, status, _ := newTestDNSPeerActivator(t)
-
-	require.NoError(t, status.AddPeer("peerA", "a.netbird.cloud", "100.64.0.1", ""))
-
-	ctx, cancel := context.WithTimeout(context.Background(), 5*time.Second)
-	defer cancel()
-
-	start := time.Now()
-	a.ActivatePeersByIP(ctx, []netip.Addr{netip.MustParseAddr("100.64.0.1")})
-	require.Less(t, time.Since(start), time.Second, "peer without a conn must not be waited on")
-}

client/internal/ebpf/ebpf/src/dns_fwd.c

@@ -52,14 +52,11 @@ int xdp_dns_fwd(struct iphdr  *ip, struct udphdr *udp) {
 
     if (udp->dest == GENERAL_DNS_PORT && ip->daddr == dns_ip) {
         udp->dest = dns_port;
-        // Clear the now-stale checksum; zero means "not computed" for IPv4.
-        udp->check = 0;
         return XDP_PASS;
     }
 
     if (udp->source == dns_port && ip->saddr == dns_ip) {
         udp->source = GENERAL_DNS_PORT;
-        udp->check = 0;
         return XDP_PASS;
     }
 

client/internal/ebpf/ebpf/src/wg_proxy.c

@@ -50,11 +50,5 @@ int xdp_wg_proxy(struct iphdr  *ip, struct udphdr *udp) {
     __be16 new_dst_port = htons(proxy_port);
     udp->dest = new_dst_port;
     udp->source = new_src_port;
-
-    // The ports are covered by the UDP checksum. This is an IPv4 loopback hop
-    // and the payload is already integrity-protected, so clear the checksum (a
-    // zero UDP checksum means "not computed" for IPv4) rather than leave a
-    // stale value the kernel would drop as UDP_CSUM.
-    udp->check = 0;
 	return XDP_PASS;
 }

client/internal/engine.go

@@ -64,10 +64,7 @@ import (
 	"github.com/netbirdio/netbird/route"
 	mgm "github.com/netbirdio/netbird/shared/management/client"
 	"github.com/netbirdio/netbird/shared/management/domain"
-	sharedgrpc "github.com/netbirdio/netbird/shared/management/grpc"
-	nbnetworkmap "github.com/netbirdio/netbird/shared/management/networkmap"
 	mgmProto "github.com/netbirdio/netbird/shared/management/proto"
-	types "github.com/netbirdio/netbird/shared/management/types"
 	"github.com/netbirdio/netbird/shared/netiputil"
 	auth "github.com/netbirdio/netbird/shared/relay/auth/hmac"
 	relayClient "github.com/netbirdio/netbird/shared/relay/client"
@@ -150,7 +147,6 @@ type EngineConfig struct {
 	BlockLANAccess      bool
 	BlockInbound        bool
 	DisableIPv6         bool
-	SyncMessageVersion  *int
 
 	// LazyConnection is the MDM-sourced lazy-connection override; StateUnset defers to
 	// the env var and management feature flag.
@@ -224,12 +220,11 @@ type Engine struct {
 	// networkSerial is the latest CurrentSerial (state ID) of the network sent by the Management service
 	networkSerial uint64
 
-	// latestComponents is the most-recent NetworkMapComponents decoded from
+	// forwardingRules holds the ingress forward rules applied for the current target.
-	// a NetworkMapEnvelope (capability=3 peers only). Held alongside the
+	// Wholesale sections (incl. forward rules) run only on the first pass of a target;
-	// NetworkMap that Calculate() produced from it so future incremental
+	// it is stashed here so the final, peer-converged pass can build the lazy-connection
-	// updates have a base to apply changes against. nil for legacy-format
+	// exclude list without recomputing them on every bounded peer pass.
-	// peers. Guarded by syncMsgMux.
+	forwardingRules []firewallManager.ForwardRule
-	latestComponents *types.NetworkMapComponents
 
 	networkMonitor *networkmonitor.NetworkMonitor
 
@@ -285,20 +280,6 @@ type Engine struct {
 	jobExecutorWG sync.WaitGroup
 
 	exposeManager *expose.Manager
-
-	sessionWatcher sessionDeadlineWatcher
-}
-
-// sessionDeadlineWatcher is the engine-facing surface of the SSO session
-// expiry watcher. The concrete implementation (sessionwatch.Watcher) is wired
-// in via newSessionWatcher, which is build-tagged so the js/wasm build links a
-// no-op stub instead of pulling the full sessionwatch package (and its timer
-// machinery) into the binary — the wasm client never runs the engine's
-// session-warning flow.
-type sessionDeadlineWatcher interface {
-	Update(deadline time.Time) error
-	Dismiss()
-	Close()
 }
 
 // Peer is an instance of the Connection Peer
@@ -350,17 +331,6 @@ func NewEngine(
 		updateManager:      services.UpdateManager,
 		syncStoreDir:       config.StateDir,
 	}
-	// sessionWatcher keeps the SubscribeStatus consumers in sync with the
-	// session expiry deadline. Deadline-change ticks come for free via
-	// Status.SetSessionExpiresAt; the watcher exists to push a wake-up at
-	// T-WarningLead and T-FinalWarningLead so the UI repaints the remaining
-	// time / warning state even when nothing else changed, and to publish
-	// two SystemEvents (the warning composition lives in sessionwatch so
-	// the wire format stays owned by one package):
-	//   - T-WarningLead   → interactive "Extend now / Dismiss" notification
-	//   - T-FinalWarningLead → auto-opened SessionAboutToExpire dialog,
-	//     suppressed when the user dismissed the earlier warning
-	engine.sessionWatcher = newSessionWatcher(engine.statusRecorder)
 
 	log.Infof("I am: %s", config.WgPrivateKey.PublicKey().String())
 	return engine
@@ -427,10 +397,6 @@ func (e *Engine) stopLocked() {
 		e.srWatcher.Close()
 	}
 
-	if e.sessionWatcher != nil {
-		e.sessionWatcher.Close()
-	}
-
 	if e.updateManager != nil {
 		e.updateManager.SetDownloadOnly()
 	}
@@ -562,7 +528,7 @@ func (e *Engine) Start(netbirdConfig *mgmProto.NetbirdConfig, mgmtURL *url.URL)
 		} else {
 			log.Infof("running rosenpass in strict mode")
 		}
-		e.rpManager, err = rosenpass.NewManager(e.config.PreSharedKey, e.config.WgIfaceName, publicKey)
+		e.rpManager, err = rosenpass.NewManager(e.config.PreSharedKey, e.config.WgIfaceName)
 		if err != nil {
 			return fmt.Errorf("create rosenpass manager: %w", err)
 		}
@@ -663,24 +629,8 @@ func (e *Engine) Start(netbirdConfig *mgmProto.NetbirdConfig, mgmtURL *url.URL)
 	iceCfg := e.createICEConfig()
 
 	e.connMgr = NewConnMgr(e.config, e.statusRecorder, e.peerStore, wgIface)
-	e.connMgr.SetRoutedIPsReconciler(func(peerKey string) error {
-		if e.routeManager == nil {
-			return nil
-		}
-		return e.routeManager.ReconcilePeerAllowedIPs(peerKey)
-	})
 	e.connMgr.Start(e.ctx)
 
-	// Wire DNS-time lazy-connection warm-up now that the connection manager
-	// exists (it does not at DNS-server construction time). A DNS answer that
-	// points at an idle peer then wakes it before the client's first request.
-	e.dnsServer.SetPeerActivator(&dnsPeerActivator{
-		connMgr:   e.connMgr,
-		peerStore: e.peerStore,
-		status:    e.statusRecorder,
-		ctx:       e.ctx,
-	})
-
 	e.srWatcher = guard.NewSRWatcher(e.signal, e.relayManager, e.mobileDep.IFaceDiscover, iceCfg)
 	e.srWatcher.Start(peer.IsForceRelayed())
 
@@ -830,7 +780,15 @@ 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
@@ -860,26 +818,32 @@ 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())
@@ -887,14 +851,19 @@ 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 {
@@ -973,72 +942,28 @@ func (e *Engine) phase(name string) func() {
 	}
 }
 
-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()
 	}
 
-	e.ApplySessionDeadline(update.GetSessionExpiresAt())
+	if update.NetworkMap != nil && update.NetworkMap.PeerConfig != nil {
-
+		e.handleAutoUpdateVersion(update.NetworkMap.PeerConfig.AutoUpdate)
-	// Envelope sync responses carry PeerConfig at the top level; legacy
-	// NetworkMap syncs carry it under NetworkMap.PeerConfig.
-	if pc := update.GetPeerConfig(); pc != nil {
-		e.handleAutoUpdateVersion(pc.GetAutoUpdate())
-	} else if nm := update.GetNetworkMap(); nm != nil && nm.GetPeerConfig() != nil {
-		e.handleAutoUpdateVersion(nm.GetPeerConfig().GetAutoUpdate())
 	}
 
 	done := e.phase("netbird_config")
 	err := e.updateNetbirdConfig(update.GetNetbirdConfig())
 	done()
 	if err != nil {
-		return err
+		return false, err
-	}
-
-	// Decode the network map from either the components envelope or the
-	// legacy proto.NetworkMap before the posture-check gating below, so the
-	// "is there a network map" decision covers both wire shapes.
-	var (
-		nm         *mgmProto.NetworkMap
-		components *types.NetworkMapComponents
-	)
-	if version := update.GetVersion(); version == int32(sharedgrpc.ComponentNetworkMap) {
-		// Components-format peer: decode the envelope back to typed
-		// components, run Calculate() locally, and convert to the wire
-		// NetworkMap shape the rest of the engine consumes. Components are
-		// retained so future incremental updates can apply deltas instead
-		// of doing a full reconstruction.
-		envelope := update.GetNetworkMapEnvelope()
-		if envelope == nil {
-			return fmt.Errorf("received a SyncReponse indicating use of components network map, but components are missing")
-		}
-
-		localKey := e.config.WgPrivateKey.PublicKey().String()
-		dnsName := ""
-		if pc := update.GetPeerConfig(); pc != nil {
-			// PeerConfig.Fqdn = "<dns_label>.<dns_domain>" — extract the
-			// shared domain by stripping the peer's own label prefix. Falls
-			// back to empty if the FQDN doesn't have the expected shape.
-			dnsName = extractDNSDomainFromFQDN(pc.GetFqdn())
-		}
-		result, err := nbnetworkmap.EnvelopeToNetworkMap(e.ctx, envelope, localKey, dnsName)
-		if err != nil {
-			return fmt.Errorf("decode network map envelope: %w", err)
-		}
-		nm = result.NetworkMap
-		components = result.Components
-	} else {
-		nm = update.GetNetworkMap()
 	}
 
 	// Posture checks are bound to the network map presence:
@@ -1046,50 +971,27 @@ func (e *Engine) handleSync(update *mgmProto.SyncResponse) error {
 	//   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
+	nm := update.GetNetworkMap()
 	if nm == nil {
-		return nil
+		return false, nil
 	}
 
 	done = e.phase("checks")
 	err = e.updateChecksIfNew(update.Checks)
 	done()
 	if err != nil {
-		return err
+		return false, err
 	}
 
-	done = e.phase("persist")
-	// Only retain the components view when the server sent the envelope
-	// path. A legacy proto.NetworkMap means components == nil; writing it
-	// here would clobber a previously-cached snapshot, breaking the
-	// incremental-delta base on a future envelope sync.
-	if components != nil {
-		e.latestComponents = components
-	}
-
-	e.persistSyncResponse(update)
-	done()
-
 	// 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 nil
+	return more, nil
-}
-
-// extractDNSDomainFromFQDN returns the trailing dotted domain part of the
-// receiving peer's FQDN — the same value the management server fills as
-// dnsName when it builds the legacy NetworkMap. "peer42.netbird.cloud" →
-// "netbird.cloud". An empty string is returned for unrecognized formats.
-func extractDNSDomainFromFQDN(fqdn string) string {
-	for i := 0; i < len(fqdn); i++ {
-		if fqdn[i] == '.' && i+1 < len(fqdn) {
-			return fqdn[i+1:]
-		}
-	}
-	return ""
 }
 
 // updateNetbirdConfig applies the management-provided NetBird configuration:
@@ -1137,6 +1039,13 @@ func (e *Engine) updateNetbirdConfig(wCfg *mgmProto.NetbirdConfig) error {
 // (not syncMsgMux) is held for the whole Set so the store cannot be cleared (disabled /
 // engine close) mid-call and have this write resurrect a file that was just removed.
 func (e *Engine) persistSyncResponse(update *mgmProto.SyncResponse) {
+	// Only persist updates that carry a network map. Config-only updates (e.g. relay
+	// token rotation, STUN/TURN) have a nil NetworkMap; persisting them would overwrite
+	// the last full map on disk and break restore-on-restart.
+	if update.GetNetworkMap() == nil {
+		return
+	}
+
 	e.syncRespMux.RLock()
 	defer e.syncRespMux.RUnlock()
 
@@ -1251,7 +1160,6 @@ func (e *Engine) applyInfoFlags(info *system.Info) {
 		e.config.BlockLANAccess,
 		e.config.BlockInbound,
 		e.config.DisableIPv6,
-		e.config.SyncMessageVersion,
 		e.config.EnableSSHRoot,
 		e.config.EnableSSHSFTP,
 		e.config.EnableSSHLocalPortForwarding,
@@ -1386,7 +1294,7 @@ func (e *Engine) handleBundle(params *mgmProto.BundleParameters) (*mgmProto.JobR
 		ClientMetrics:  e.clientMetrics,
 		DaemonVersion:  version.NetbirdVersion(),
 		RefreshStatus: func() {
-			e.RunHealthProbes(e.ctx, true)
+			e.RunHealthProbes(true)
 		},
 	}
 
@@ -1425,7 +1333,24 @@ func (e *Engine) receiveManagementEvents() {
 		}
 		e.applyInfoFlags(info)
 
-		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.
+		persist := func(u *mgmProto.SyncResponse) {
+			done := e.phase("persist")
+			e.persistSyncResponse(u)
+			done()
+		}
+		manager := newMapStateManager(e.applySyncPass, persist, 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
@@ -1476,21 +1401,107 @@ 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()))
+
+	doneOffline := e.phase("offline_peers")
+	e.updateOfflinePeers(networkMap.GetOfflinePeers())
+	doneOffline()
+
+	// 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)
+		}
+	}
+
+	// No special case for cleanup: when management signals RemotePeersIsEmpty (e.g. our
+	// peer was deleted), remotePeers is already empty, so the bounded diff below removes
+	// every peer in batches — same path as a normal update, no unbounded removeAllPeers
+	// held under syncMsgMux in one shot.
+	doneRemoved := e.phase("removed_peers")
+	removeMore, err := e.removePeers(remotePeers, maxBatch)
+	doneRemoved()
+	if err != nil {
+		return false, err
+	}
+
+	doneModified := e.phase("modified_peers")
+	modifyMore, err := e.modifyPeers(remotePeers, maxBatch)
+	doneModified()
+	if err != nil {
+		return false, err
+	}
+
+	doneAdded := e.phase("added_peers")
+	addMore, err := e.addNewPeers(remotePeers, maxBatch)
+	doneAdded()
+	if err != nil {
+		return false, err
+	}
+
+	// needMore signals the caller to re-run when a peer phase hit its per-pass cap.
+	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 {
+		doneLazy := e.phase("lazy_exclude")
+		excludedLazyPeers := e.toExcludedLazyPeers(e.forwardingRules, remotePeers)
+		e.connMgr.SetExcludeList(e.ctx, excludedLazyPeers)
+		doneLazy()
+	}
+
+	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)
 	}
@@ -1563,84 +1574,7 @@ func (e *Engine) updateNetworkMap(networkMap *mgmProto.NetworkMap) error {
 		log.Errorf("failed to update forward rules, err: %v", err)
 	}
 	done()
-
+	e.forwardingRules = forwardingRules
-	log.Debugf("got peers update from Management Service, total peers to connect to = %d", len(networkMap.GetRemotePeers()))
-
-	done = e.phase("offline_peers")
-	e.updateOfflinePeers(networkMap.GetOfflinePeers())
-	done()
-
-	remotePeers, err := e.reconcilePeers(networkMap)
-	if err != nil {
-		return err
-	}
-
-	// must set the exclude list after the peers are added. Without it the manager can not figure out the peers parameters from the store
-	done = e.phase("lazy_exclude")
-	excludedLazyPeers := e.toExcludedLazyPeers(forwardingRules, remotePeers)
-	e.connMgr.SetExcludeList(e.ctx, excludedLazyPeers)
-	done()
-
-	e.networkSerial = serial
-
-	return nil
-}
-
-// reconcilePeers applies the remote peer list from the network map (removing,
-// modifying and adding peers, then updating SSH config) and returns the remote
-// peers with our own peer filtered out, for use by later sync steps.
-func (e *Engine) reconcilePeers(networkMap *mgmProto.NetworkMap) ([]*mgmProto.RemotePeerConfig, error) {
-	// 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 nil, err
-		}
-		return remotePeers, nil
-	}
-
-	done := e.phase("removed_peers")
-	err := e.removePeers(remotePeers)
-	done()
-	if err != nil {
-		return nil, err
-	}
-
-	done = e.phase("modified_peers")
-	err = e.modifyPeers(remotePeers)
-	done()
-	if err != nil {
-		return nil, err
-	}
-
-	done = e.phase("added_peers")
-	err = e.addNewPeers(remotePeers)
-	done()
-	if err != nil {
-		return nil, 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())
-
-	return remotePeers, nil
 }
 
 func toDNSFeatureFlag(networkMap *mgmProto.NetworkMap) bool {
@@ -1820,14 +1754,23 @@ 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
@@ -2120,7 +2063,6 @@ func (e *Engine) readInitialSettings() ([]*route.Route, *nbdns.Config, bool, err
 		e.config.BlockLANAccess,
 		e.config.BlockInbound,
 		e.config.DisableIPv6,
-		e.config.SyncMessageVersion,
 		e.config.EnableSSHRoot,
 		e.config.EnableSSHSFTP,
 		e.config.EnableSSHLocalPortForwarding,
@@ -2313,20 +2255,7 @@ func (e *Engine) getRosenpassAddr() string {
 
 // RunHealthProbes executes health checks for Signal, Management, Relay, and WireGuard services
 // and updates the status recorder with the latest states.
-//
+func (e *Engine) RunHealthProbes(waitForResult bool) bool {
-// ctx scopes the (potentially slow) STUN/TURN probing: a caller that gives up —
-// e.g. a Status RPC whose client disconnected — cancels its ctx and the probe
-// returns instead of running to its per-component timeout. The engine's own
-// lifetime ctx still applies independently, so an engine shutdown aborts the
-// probe even if the caller's ctx is context.Background().
-func (e *Engine) RunHealthProbes(ctx context.Context, waitForResult bool) bool {
-	// Tie the caller's ctx to the engine lifetime: either cancelling aborts
-	// the probe below.
-	ctx, cancel := context.WithCancel(ctx)
-	defer cancel()
-	stop := context.AfterFunc(e.ctx, cancel)
-	defer stop()
-
 	e.syncMsgMux.Lock()
 
 	signalHealthy := e.signal.IsHealthy()
@@ -2349,9 +2278,9 @@ func (e *Engine) RunHealthProbes(ctx context.Context, waitForResult bool) bool {
 	if runtime.GOOS != "js" {
 		var results []relay.ProbeResult
 		if waitForResult {
-			results = e.probeStunTurn.ProbeAllWaitResult(ctx, stuns, turns)
+			results = e.probeStunTurn.ProbeAllWaitResult(e.ctx, stuns, turns)
 		} else {
-			results = e.probeStunTurn.ProbeAll(ctx, stuns, turns)
+			results = e.probeStunTurn.ProbeAll(e.ctx, stuns, turns)
 		}
 		e.statusRecorder.UpdateRelayStates(results)
 
@@ -2694,14 +2623,13 @@ func (e *Engine) updateForwardRules(rules []*mgmProto.ForwardingRule) ([]firewal
 
 func (e *Engine) toExcludedLazyPeers(rules []firewallManager.ForwardRule, peers []*mgmProto.RemotePeerConfig) map[string]bool {
 	excludedPeers := make(map[string]bool)
-
-	// Ingress forward targets: inbound forwarded traffic is initiated remotely and
-	// cannot wake a lazy connection, so the peer routing the target must stay
-	// permanently connected. AllowedIPs are already parsed on the peer conn, so
-	// reuse those typed prefixes instead of re-parsing the network map strings.
 	for _, r := range rules {
+		ip := r.TranslatedAddress
 		for _, p := range peers {
-			if e.peerRoutesAddr(p, r.TranslatedAddress) {
+			for _, allowedIP := range p.GetAllowedIps() {
+				if allowedIP != ip.String() {
+					continue
+				}
 				log.Infof("exclude forwarder peer from lazy connection: %s", p.GetWgPubKey())
 				excludedPeers[p.GetWgPubKey()] = true
 			}
@@ -2711,27 +2639,6 @@ func (e *Engine) toExcludedLazyPeers(rules []firewallManager.ForwardRule, peers
 	return excludedPeers
 }
 
-// peerRoutesAddr reports whether the peer is a router for addr, matched against
-// the peer's already-parsed AllowedIPs from the store (the same typed value the
-// lazy manager consumes) rather than re-parsing the network map strings.
-func (e *Engine) peerRoutesAddr(p *mgmProto.RemotePeerConfig, addr netip.Addr) bool {
-	prefixes, ok := e.peerStore.AllowedIPs(p.GetWgPubKey())
-	if !ok {
-		return false
-	}
-	return prefixesContain(prefixes, addr)
-}
-
-// prefixesContain reports whether addr falls within any of the prefixes.
-func prefixesContain(prefixes []netip.Prefix, addr netip.Addr) bool {
-	for _, prefix := range prefixes {
-		if prefix.Contains(addr) {
-			return true
-		}
-	}
-	return false
-}
-
 // isChecksEqual checks if two slices of checks are equal.
 func isChecksEqual(checks1, checks2 []*mgmProto.Checks) bool {
 	normalize := func(checks []*mgmProto.Checks) []string {

client/internal/engine_authsession.go

@@ -1,108 +0,0 @@
-package internal
-
-import (
-	"context"
-	"errors"
-	"fmt"
-	"time"
-
-	log "github.com/sirupsen/logrus"
-	"google.golang.org/protobuf/types/known/timestamppb"
-
-	"github.com/netbirdio/netbird/client/internal/auth/sessionwatch"
-	cProto "github.com/netbirdio/netbird/client/proto"
-	"github.com/netbirdio/netbird/client/system"
-)
-
-// ApplySessionDeadline propagates the absolute SSO session deadline carried on
-// LoginResponse / SyncResponse to both the watcher (for the edge-triggered
-// warning) and the status recorder (for the SubscribeStatus / Status RPC
-// snapshot the UI consumes).
-//
-// The wire field is 3-state:
-//   - nil                        → snapshot carries no info; keep the
-//     previously-anchored deadline (no-op)
-//   - explicit zero (s=0, n=0)   → peer is not SSO-registered or expiry is
-//     disabled; clear both sinks
-//   - valid timestamp            → new deadline; arm watcher, expose on
-//     status recorder
-//
-// Deadline sanity-checks live in sessionwatch.Watcher.Update. Any rejected
-// value is treated as a clear on both sinks: the alternative — leaving the
-// previously-known deadline in place — risks the UI confidently displaying
-// a stale "expires in X" while the server has actually invalidated it.
-func (e *Engine) ApplySessionDeadline(ts *timestamppb.Timestamp) {
-	if ts == nil {
-		return
-	}
-	var deadline time.Time
-	// Explicit zero (seconds=0 AND nanos=0) is the sentinel for "disabled".
-	// Everything else flows through Watcher.Update, whose sanity-checks
-	// reject out-of-range / pre-epoch / far-future / too-stale values and
-	// clear on rejection.
-	if ts.GetSeconds() != 0 || ts.GetNanos() != 0 {
-		deadline = ts.AsTime().UTC()
-	}
-	if e.sessionWatcher == nil {
-		return
-	}
-	// Watcher.Update owns the propagation to the status recorder (the
-	// SubscribeStatus / Status snapshot the UI reads): a set writes the
-	// deadline, a clear or a sanity-check rejection writes the zero value.
-	// Keeping a single writer is what stops the recorder from drifting out
-	// of sync with the warning timers.
-	if err := e.sessionWatcher.Update(deadline); err != nil {
-		log.Errorf("auth session deadline rejected: %v, clearing", err)
-		e.statusRecorder.PublishEvent(
-			cProto.SystemEvent_ERROR,
-			cProto.SystemEvent_AUTHENTICATION,
-			"session deadline rejected",
-			"",
-			map[string]string{sessionwatch.MetaSessionDeadlineRejected: err.Error()},
-		)
-	}
-}
-
-// DismissSessionWarning records the user's "Dismiss" click on the
-// T-WarningLead interactive notification and suppresses the upcoming
-// T-FinalWarningLead fallback for the current deadline. No-op when the
-// watcher is not running or holds no deadline.
-func (e *Engine) DismissSessionWarning() {
-	if e.sessionWatcher == nil {
-		return
-	}
-	e.sessionWatcher.Dismiss()
-}
-
-// ExtendAuthSession asks the management server to refresh the SSO session
-// expiry deadline using the supplied JWT, then mirrors the new deadline into
-// the daemon's state. The tunnel is untouched; no resync, no reconnect.
-//
-// Returns the new absolute UTC deadline (or zero time when the server
-// reports the peer is not eligible for extension).
-func (e *Engine) ExtendAuthSession(ctx context.Context, jwtToken string) (time.Time, error) {
-	if jwtToken == "" {
-		return time.Time{}, errors.New("jwt token is required")
-	}
-	if e.mgmClient == nil {
-		return time.Time{}, errors.New("management client is not initialised")
-	}
-
-	info, err := system.GetInfoWithChecks(ctx, e.checks)
-	if err != nil {
-		log.Warnf("failed to collect system info for session extend: %v", err)
-		info = system.GetInfo(ctx)
-	}
-
-	resp, err := e.mgmClient.ExtendAuthSession(info, jwtToken)
-	if err != nil {
-		return time.Time{}, fmt.Errorf("extend auth session on management: %w", err)
-	}
-
-	e.ApplySessionDeadline(resp.GetSessionExpiresAt())
-
-	if resp.GetSessionExpiresAt().IsValid() {
-		return resp.GetSessionExpiresAt().AsTime().UTC(), nil
-	}
-	return time.Time{}, nil
-}

client/internal/engine_lazy_exclude_test.go

@@ -1,87 +0,0 @@
-package internal
-
-import (
-	"net/netip"
-	"testing"
-
-	"github.com/stretchr/testify/require"
-
-	firewallManager "github.com/netbirdio/netbird/client/firewall/manager"
-	"github.com/netbirdio/netbird/client/internal/peer"
-	"github.com/netbirdio/netbird/client/internal/peerstore"
-	mgmProto "github.com/netbirdio/netbird/shared/management/proto"
-)
-
-func TestPrefixesContain(t *testing.T) {
-	tests := []struct {
-		name     string
-		prefixes []string
-		addr     string
-		want     bool
-	}{
-		{name: "own overlay /32 matches", prefixes: []string{"100.110.8.145/32"}, addr: "100.110.8.145", want: true},
-		{name: "addr inside routed subnet", prefixes: []string{"10.121.0.0/16"}, addr: "10.121.208.4", want: true},
-		{name: "addr outside subnet", prefixes: []string{"10.121.0.0/16"}, addr: "10.122.0.1", want: false},
-		{name: "different /32", prefixes: []string{"100.110.8.145/32"}, addr: "100.110.8.146", want: false},
-		{name: "ipv6 /128 matches", prefixes: []string{"fd00::1/128"}, addr: "fd00::1", want: true},
-		{name: "no prefixes", prefixes: nil, addr: "10.121.208.4", want: false},
-	}
-
-	for _, tt := range tests {
-		t.Run(tt.name, func(t *testing.T) {
-			prefixes := make([]netip.Prefix, 0, len(tt.prefixes))
-			for _, p := range tt.prefixes {
-				prefixes = append(prefixes, netip.MustParsePrefix(p))
-			}
-			require.Equal(t, tt.want, prefixesContain(prefixes, netip.MustParseAddr(tt.addr)))
-		})
-	}
-}
-
-// TestToExcludedLazyPeers_ForwardTarget guards a regression: the forward-target
-// peer (the peer routing a ForwardRule.TranslatedAddress) must be excluded from
-// lazy connections, matched via the peer's already-parsed AllowedIPs.
-func TestToExcludedLazyPeers_ForwardTarget(t *testing.T) {
-	const targetPeerKey = "cccccccccccccccccccccccccccccccccccccccccc0="
-	const otherPeerKey = "dddddddddddddddddddddddddddddddddddddddddd0="
-
-	store := peerstore.NewConnStore()
-	store.AddPeerConn(targetPeerKey, newTestConn(t, targetPeerKey, "100.110.8.145/32"))
-	store.AddPeerConn(otherPeerKey, newTestConn(t, otherPeerKey, "100.110.9.10/32"))
-
-	e := &Engine{peerStore: store}
-
-	peers := []*mgmProto.RemotePeerConfig{
-		{WgPubKey: targetPeerKey, AllowedIps: []string{"100.110.8.145/32"}},
-		{WgPubKey: otherPeerKey, AllowedIps: []string{"100.110.9.10/32"}},
-	}
-	rules := []firewallManager.ForwardRule{
-		{TranslatedAddress: netip.MustParseAddr("100.110.8.145")},
-	}
-
-	excluded := e.toExcludedLazyPeers(rules, peers)
-
-	require.True(t, excluded[targetPeerKey], "forward-target peer must be excluded from lazy connections")
-	require.False(t, excluded[otherPeerKey], "non-target peer must not be excluded")
-	require.Len(t, excluded, 1)
-}
-
-func TestToExcludedLazyPeers_NoRules(t *testing.T) {
-	e := &Engine{peerStore: peerstore.NewConnStore()}
-
-	peers := []*mgmProto.RemotePeerConfig{
-		{WgPubKey: "peer-a", AllowedIps: []string{"100.110.8.145/32"}},
-	}
-
-	require.Empty(t, e.toExcludedLazyPeers(nil, peers))
-}
-
-func newTestConn(t *testing.T, key, allowedIP string) *peer.Conn {
-	t.Helper()
-	conn, err := peer.NewConn(peer.ConnConfig{
-		Key:      key,
-		WgConfig: peer.WgConfig{AllowedIps: []netip.Prefix{netip.MustParsePrefix(allowedIP)}},
-	}, peer.ServiceDependencies{})
-	require.NoError(t, err)
-	return conn
-}

client/internal/engine_privileged_test.go

@@ -124,7 +124,7 @@ func TestEngine_SSH(t *testing.T) {
 		RemotePeersIsEmpty: false,
 	}
 
-	err = engine.updateNetworkMap(networkMap)
+	_, err = engine.updateNetworkMap(networkMap, maxPeersPerSyncPass, true)
 	require.NoError(t, err)
 
 	assert.Nil(t, engine.sshServer)
@@ -146,7 +146,7 @@ func TestEngine_SSH(t *testing.T) {
 		RemotePeersIsEmpty: false,
 	}
 
-	err = engine.updateNetworkMap(networkMap)
+	_, err = engine.updateNetworkMap(networkMap, maxPeersPerSyncPass, true)
 	require.NoError(t, err)
 
 	time.Sleep(250 * time.Millisecond)
@@ -159,7 +159,7 @@ func TestEngine_SSH(t *testing.T) {
 		RemotePeersIsEmpty: false,
 	}
 
-	err = engine.updateNetworkMap(networkMap)
+	_, err = engine.updateNetworkMap(networkMap, maxPeersPerSyncPass, true)
 	require.NoError(t, err)
 
 	// time.Sleep(250 * time.Millisecond)
@@ -174,7 +174,7 @@ func TestEngine_SSH(t *testing.T) {
 		RemotePeersIsEmpty: false,
 	}
 
-	err = engine.updateNetworkMap(networkMap)
+	_, err = engine.updateNetworkMap(networkMap, maxPeersPerSyncPass, true)
 	require.NoError(t, err)
 
 	assert.Nil(t, engine.sshServer)

client/internal/engine_session_deadline_test.go

@@ -1,88 +0,0 @@
-package internal
-
-import (
-	"testing"
-	"time"
-
-	"github.com/stretchr/testify/require"
-	"google.golang.org/protobuf/types/known/timestamppb"
-
-	"github.com/netbirdio/netbird/client/internal/auth/sessionwatch"
-	"github.com/netbirdio/netbird/client/internal/peer"
-)
-
-// TestApplySessionDeadline_ThreeState pins down the 3-state semantics of the
-// wire field carried on LoginResponse / SyncResponse:
-//
-//   - nil pointer            → no info; previously-anchored deadline survives
-//   - explicit zero value    → "expiry disabled" sentinel; both sinks cleared
-//   - valid future timestamp → new deadline propagated to both sinks
-func TestApplySessionDeadline_ThreeState(t *testing.T) {
-	newEngine := func() *Engine {
-		recorder := peer.NewRecorder("")
-		return &Engine{
-			statusRecorder: recorder,
-			sessionWatcher: sessionwatch.New(recorder),
-		}
-	}
-
-	t.Run("valid timestamp sets deadline on both sinks", func(t *testing.T) {
-		e := newEngine()
-		deadline := time.Now().Add(time.Hour).UTC().Truncate(time.Second)
-
-		e.ApplySessionDeadline(timestamppb.New(deadline))
-
-		require.True(t, e.statusRecorder.GetSessionExpiresAt().Equal(deadline),
-			"status recorder should hold the new deadline")
-	})
-
-	t.Run("nil is a no-op and preserves previous deadline", func(t *testing.T) {
-		e := newEngine()
-		seeded := time.Now().Add(time.Hour).UTC().Truncate(time.Second)
-		e.ApplySessionDeadline(timestamppb.New(seeded))
-		require.True(t, e.statusRecorder.GetSessionExpiresAt().Equal(seeded))
-
-		e.ApplySessionDeadline(nil)
-
-		require.True(t, e.statusRecorder.GetSessionExpiresAt().Equal(seeded),
-			"nil snapshot must not disturb the existing deadline")
-	})
-
-	t.Run("explicit zero clears a previously-anchored deadline", func(t *testing.T) {
-		e := newEngine()
-		seeded := time.Now().Add(time.Hour).UTC().Truncate(time.Second)
-		e.ApplySessionDeadline(timestamppb.New(seeded))
-		require.True(t, e.statusRecorder.GetSessionExpiresAt().Equal(seeded))
-
-		// Explicit zero Timestamp{} (seconds=0, nanos=0) is the
-		// "expiry disabled / not SSO" sentinel.
-		e.ApplySessionDeadline(&timestamppb.Timestamp{})
-
-		require.True(t, e.statusRecorder.GetSessionExpiresAt().IsZero(),
-			"explicit zero sentinel must clear the deadline")
-	})
-
-	t.Run("invalid timestamp clears the deadline", func(t *testing.T) {
-		e := newEngine()
-		seeded := time.Now().Add(time.Hour).UTC().Truncate(time.Second)
-		e.ApplySessionDeadline(timestamppb.New(seeded))
-		require.True(t, e.statusRecorder.GetSessionExpiresAt().Equal(seeded))
-
-		// Out-of-range nanos → IsValid()==false; same-meaning as the
-		// disabled sentinel for downstream sinks.
-		e.ApplySessionDeadline(&timestamppb.Timestamp{Seconds: 1, Nanos: -1})
-
-		require.True(t, e.statusRecorder.GetSessionExpiresAt().IsZero(),
-			"invalid timestamp must clear the deadline")
-	})
-
-	t.Run("recently expired timestamp stays visible as expired", func(t *testing.T) {
-		e := newEngine()
-		expired := time.Now().Add(-5 * time.Minute).UTC().Truncate(time.Second)
-
-		e.ApplySessionDeadline(timestamppb.New(expired))
-
-		require.True(t, e.statusRecorder.GetSessionExpiresAt().Equal(expired),
-			"recently-expired deadline must stay on the recorder so consumers render it as expired")
-	})
-}

client/internal/engine_sessionwatch.go

@@ -1,16 +0,0 @@
-//go:build !js
-
-package internal
-
-import (
-	"github.com/netbirdio/netbird/client/internal/auth/sessionwatch"
-	"github.com/netbirdio/netbird/client/internal/peer"
-)
-
-// newSessionWatcher returns the real SSO session expiry watcher for every
-// non-wasm build. The js/wasm build gets a no-op stub from
-// engine_sessionwatch_js.go so the sessionwatch package (and its timer
-// machinery) never links into the wasm binary.
-func newSessionWatcher(recorder *peer.Status) sessionDeadlineWatcher {
-	return sessionwatch.New(recorder)
-}

client/internal/engine_sessionwatch_js.go

@@ -1,44 +0,0 @@
-//go:build js
-
-package internal
-
-import (
-	"time"
-
-	"github.com/netbirdio/netbird/client/internal/peer"
-)
-
-// noopSessionWatcher is the js/wasm stand-in for sessionwatch.Watcher. The
-// wasm client never runs the engine's session-warning flow (the interactive
-// T-WarningLead notification and the T-FinalWarningLead fallback dialog live
-// in the desktop UI), so linking the full sessionwatch package (timers, event
-// composition) would only bloat the binary.
-//
-// It still mirrors the deadline into the status recorder so the SubscribeStatus
-// / Status snapshot the UI consumes stays correct — only the timer-driven
-// warnings are dropped.
-type noopSessionWatcher struct {
-	recorder *peer.Status
-}
-
-func newSessionWatcher(recorder *peer.Status) sessionDeadlineWatcher {
-	return noopSessionWatcher{recorder: recorder}
-}
-
-// Update mirrors the real watcher's recorder propagation without the timers or
-// sanity-check sentinels: a valid deadline is exposed on the status snapshot,
-// the zero time clears it.
-func (w noopSessionWatcher) Update(deadline time.Time) error {
-	if w.recorder != nil {
-		w.recorder.SetSessionExpiresAt(deadline)
-	}
-	return nil
-}
-
-func (noopSessionWatcher) Dismiss() {
-	// No-op: only suppresses the timer-driven final-warning, which this stub never arms.
-}
-
-func (noopSessionWatcher) Close() {
-	// No-op: no timers to stop and no state to unwind; the recorder is cleared via Update(zero).
-}

client/internal/engine_test.go

@@ -437,7 +437,7 @@ 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
@@ -464,6 +464,47 @@ func TestEngine_UpdateNetworkMap(t *testing.T) {
 			}
 		})
 	}
+
+	// chunked apply: with a per-pass cap smaller than the number of peers, a
+	// single updateNetworkMap applies one batch and reports more==true; the
+	// caller re-runs until convergence. (engine currently holds 0 peers.)
+	t.Run("chunked add converges over multiple passes", func(t *testing.T) {
+		nm := &mgmtProto.NetworkMap{
+			Serial:      6,
+			RemotePeers: []*mgmtProto.RemotePeerConfig{peer1, peer2, peer3},
+		}
+
+		more, err := engine.updateNetworkMap(nm, 1, true)
+		require.NoError(t, err)
+		require.True(t, more, "pass 1 should signal more")
+		require.Len(t, engine.peerStore.PeersPubKey(), 1)
+
+		more, err = engine.updateNetworkMap(nm, 1, false)
+		require.NoError(t, err)
+		require.True(t, more, "pass 2 should signal more")
+		require.Len(t, engine.peerStore.PeersPubKey(), 2)
+
+		more, err = engine.updateNetworkMap(nm, 1, false)
+		require.NoError(t, err)
+		require.False(t, more, "pass 3 should converge")
+		require.Len(t, engine.peerStore.PeersPubKey(), 3)
+	})
+
+	t.Run("chunked remove converges over multiple passes", func(t *testing.T) {
+		nm := &mgmtProto.NetworkMap{
+			Serial:      7,
+			RemotePeers: []*mgmtProto.RemotePeerConfig{peer1}, // remove peer2, peer3
+		}
+
+		more, err := engine.updateNetworkMap(nm, 1, true)
+		require.NoError(t, err)
+		require.True(t, more, "pass 1 should signal more (2 to remove, cap 1)")
+
+		more, err = engine.updateNetworkMap(nm, 1, false)
+		require.NoError(t, err)
+		require.False(t, more, "pass 2 should converge")
+		require.Len(t, engine.peerStore.PeersPubKey(), 1)
+	})
 }
 
 func TestEngine_UpdateNetworkMapWithRoutes(t *testing.T) {
@@ -634,7 +675,7 @@ 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")
@@ -838,7 +879,7 @@ 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")

client/internal/lazyconn/manager/manager.go

@@ -29,11 +29,6 @@ type managedPeer struct {
 
 type Config struct {
 	InactivityThreshold *time.Duration
-	// ReconcileAllowedIPs re-applies a peer's routed allowed IPs after its wake endpoint is
-	// armed. The activity listener creates the wake peer with the overlay /32 only; without the
-	// routed prefixes WireGuard would not steer subnet-bound traffic to the wake endpoint, so an
-	// idle routing peer could never be woken by that traffic. Optional; nil disables the reconcile.
-	ReconcileAllowedIPs func(peerKey string) error
 }
 
 // Manager manages lazy connections
@@ -61,9 +56,6 @@ type Manager struct {
 	peerToHAGroups map[string][]route.HAUniqueID // peer ID -> HA groups they belong to
 	haGroupToPeers map[route.HAUniqueID][]string // HA group -> peer IDs in the group
 	routesMu       sync.RWMutex
-
-	// reconcileAllowedIPs re-applies a peer's routed allowed IPs after its wake endpoint is armed.
-	reconcileAllowedIPs func(peerKey string) error
 }
 
 // NewManager creates a new lazy connection manager
@@ -81,7 +73,6 @@ func NewManager(config Config, engineCtx context.Context, peerStore *peerstore.S
 		activityManager:      activity.NewManager(wgIface),
 		peerToHAGroups:       make(map[string][]route.HAUniqueID),
 		haGroupToPeers:       make(map[route.HAUniqueID][]string),
-		reconcileAllowedIPs:  config.ReconcileAllowedIPs,
 	}
 
 	if wgIface.IsUserspaceBind() {
@@ -210,7 +201,7 @@ func (m *Manager) AddPeer(peerCfg lazyconn.PeerConfig) (bool, error) {
 		return false, nil
 	}
 
-	if err := m.armActivityListener(peerCfg); err != nil {
+	if err := m.activityManager.MonitorPeerActivity(peerCfg); err != nil {
 		return false, err
 	}
 
@@ -297,7 +288,7 @@ func (m *Manager) DeactivatePeer(peerID peerid.ConnID) {
 
 	m.inactivityManager.RemovePeer(mp.peerCfg.PublicKey)
 
-	if err := m.armActivityListener(*mp.peerCfg); err != nil {
+	if err := m.activityManager.MonitorPeerActivity(*mp.peerCfg); err != nil {
 		mp.peerCfg.Log.Errorf("failed to create activity monitor: %v", err)
 		return
 	}
@@ -474,31 +465,6 @@ func (m *Manager) close() {
 }
 
 // shouldDeferIdleForHA checks if peer should stay connected due to HA group requirements
-// armRoutedAllowedIPs re-applies the peer's routed allowed IPs onto its freshly armed wake
-// endpoint. The activity listener creates the wake peer with the overlay /32 only, so without
-// this the routed prefixes would be missing and traffic to a routed subnet could not wake the
-// idle routing peer. It is a no-op when no reconciler is configured.
-// armActivityListener (re)arms the peer's wake endpoint via the activity manager and then
-// re-applies its routed allowed IPs, so traffic to a routed subnet can wake an idle routing
-// peer. The routed prefixes must be re-applied after the wake endpoint exists because the
-// listener creates it with the overlay /32 only.
-func (m *Manager) armActivityListener(peerCfg lazyconn.PeerConfig) error {
-	if err := m.activityManager.MonitorPeerActivity(peerCfg); err != nil {
-		return err
-	}
-	m.armRoutedAllowedIPs(&peerCfg)
-	return nil
-}
-
-func (m *Manager) armRoutedAllowedIPs(peerCfg *lazyconn.PeerConfig) {
-	if m.reconcileAllowedIPs == nil {
-		return
-	}
-	if err := m.reconcileAllowedIPs(peerCfg.PublicKey); err != nil {
-		peerCfg.Log.Errorf("failed to reconcile routed allowed IPs on wake endpoint: %v", err)
-	}
-}
-
 func (m *Manager) shouldDeferIdleForHA(inactivePeers map[string]struct{}, peerID string) bool {
 	m.routesMu.RLock()
 	defer m.routesMu.RUnlock()
@@ -611,7 +577,7 @@ func (m *Manager) onPeerInactivityTimedOut(peerIDs map[string]struct{}) {
 
 		mp.peerCfg.Log.Infof("start activity monitor")
 
-		if err := m.armActivityListener(*mp.peerCfg); err != nil {
+		if err := m.activityManager.MonitorPeerActivity(*mp.peerCfg); err != nil {
 			mp.peerCfg.Log.Errorf("failed to create activity monitor: %v", err)
 			continue
 		}

client/internal/mapsync.go

@@ -0,0 +1,214 @@
+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 {
+	// Nothing pending to preserve (no prev, or prev already fully applied): plain replace.
+	if prev == nil || update == nil || m.targetGen == m.appliedGen {
+		return update
+	}
+
+	// prev still has unapplied state (targetGen > appliedGen). In the sync protocol a
+	// nil component means "no change", so if `update` omits a component that prev
+	// carried, carry prev's forward — otherwise coalescing an update that superseded a
+	// not-yet-applied one would silently drop the map or config it uniquely brought.
+	// A present component in `update` is newer and wins. Management may send map-only
+	// updates (nil config) and config-only updates (nil map); both are handled here.
+	// A nil component in `update` means "no change", so fill it in from prev — otherwise
+	// coalescing an update that superseded a not-yet-applied one would drop the map or
+	// config it uniquely carried. A present component in `update` is newer and wins.
+	// We mutate `update` in place: it is a fresh per-message allocation from the sync
+	// stream (see receiveUpdatesEvents — not reused), and persisting this squashed target
+	// is correct, since it is the current full (superset) desired state.
+	if update.GetNetworkMap() == nil && prev.GetNetworkMap() != nil {
+		update.NetworkMap = prev.GetNetworkMap()
+		update.Checks = prev.Checks // checks travel with the map
+	}
+	if update.GetNetbirdConfig() == nil && prev.GetNetbirdConfig() != nil {
+		update.NetbirdConfig = prev.GetNetbirdConfig()
+	}
+	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 +1,281 @@
+package internal
+
+import (
+	"context"
+	"errors"
+	"sync/atomic"
+	"testing"
+	"time"
+
+	"github.com/stretchr/testify/require"
+
+	mgmProto "github.com/netbirdio/netbird/shared/management/proto"
+)
+
+// mergeTarget fills components missing from the incoming update with the pending
+// (not-yet-applied) prev's, in place, so a coalesced/superseded update does not drop
+// the map or config it uniquely carried.
+func TestMapStateManager_MergeTargetPreservesPendingState(t *testing.T) {
+	m := newMapStateManager(nil, nil, nil)
+
+	// config-only update while a full map is still converging (targetGen > appliedGen):
+	// the pending map (+ checks) is filled into the update in place
+	m.targetGen, m.appliedGen = 5, 4
+	prev := &mgmProto.SyncResponse{NetworkMap: &mgmProto.NetworkMap{Serial: 5}}
+	update := &mgmProto.SyncResponse{NetbirdConfig: &mgmProto.NetbirdConfig{}}
+	merged := m.mergeTarget(prev, update)
+	require.Same(t, update, merged, "merges in place, returns the update")
+	require.EqualValues(t, 5, merged.GetNetworkMap().GetSerial(), "pending map preserved")
+	require.NotNil(t, merged.GetNetbirdConfig(), "new config kept")
+
+	// symmetric: map-only update while a config-only update is pending -> keep the config
+	m.targetGen, m.appliedGen = 5, 4
+	prev = &mgmProto.SyncResponse{NetbirdConfig: &mgmProto.NetbirdConfig{}}
+	update = &mgmProto.SyncResponse{NetworkMap: &mgmProto.NetworkMap{Serial: 7}}
+	merged = m.mergeTarget(prev, update)
+	require.EqualValues(t, 7, merged.GetNetworkMap().GetSerial(), "new map kept")
+	require.NotNil(t, merged.GetNetbirdConfig(), "pending config preserved")
+
+	// prev already applied (targetGen == appliedGen): plain replace, no fill-in
+	m.targetGen, m.appliedGen = 5, 5
+	prev = &mgmProto.SyncResponse{NetworkMap: &mgmProto.NetworkMap{Serial: 5}}
+	update = &mgmProto.SyncResponse{NetbirdConfig: &mgmProto.NetbirdConfig{}}
+	merged = m.mergeTarget(prev, update)
+	require.Same(t, update, merged)
+	require.Nil(t, merged.GetNetworkMap(), "no map grafted when prev already applied")
+
+	// nothing to carry (update has a map, prev has no config): plain replace
+	m.targetGen, m.appliedGen = 5, 4
+	prev = &mgmProto.SyncResponse{NetworkMap: &mgmProto.NetworkMap{Serial: 5}}
+	update = &mgmProto.SyncResponse{NetworkMap: &mgmProto.NetworkMap{Serial: 6}}
+	require.Same(t, update, m.mergeTarget(prev, update))
+}
+
+// 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/netflow/logger/logger.go

@@ -27,7 +27,7 @@ type Logger struct {
 	wgIfaceNetV6       netip.Prefix
 	dnsCollection      atomic.Bool
 	exitNodeCollection atomic.Bool
-	Store              types.AggregatingStore
+	Store              types.Store
 }
 
 func New(statusRecorder *peer.Status, wgIfaceIPNet, wgIfaceIPNetV6 netip.Prefix) *Logger {
@@ -35,7 +35,7 @@ func New(statusRecorder *peer.Status, wgIfaceIPNet, wgIfaceIPNetV6 netip.Prefix)
 		statusRecorder: statusRecorder,
 		wgIfaceNet:     wgIfaceIPNet,
 		wgIfaceNetV6:   wgIfaceIPNetV6,
-		Store:          store.NewAggregatingMemoryStore(),
+		Store:          store.NewMemoryStore(),
 	}
 }
 
@@ -125,10 +125,6 @@ func (l *Logger) stop() {
 	l.mux.Unlock()
 }
 
-func (l *Logger) ResetAggregationWindow() types.FlowEventAggregator {
-	return l.Store.ResetAggregationWindow()
-}
-
 func (l *Logger) GetEvents() []*types.Event {
 	return l.Store.GetEvents()
 }

client/internal/netflow/manager.go

@@ -9,14 +9,12 @@ import (
 	"sync"
 	"time"
 
-	"github.com/cenkalti/backoff/v4"
 	"github.com/google/uuid"
 	log "github.com/sirupsen/logrus"
 	"google.golang.org/protobuf/types/known/timestamppb"
 
 	"github.com/netbirdio/netbird/client/internal/netflow/conntrack"
 	"github.com/netbirdio/netbird/client/internal/netflow/logger"
-	"github.com/netbirdio/netbird/client/internal/netflow/store"
 	nftypes "github.com/netbirdio/netbird/client/internal/netflow/types"
 	"github.com/netbirdio/netbird/client/internal/peer"
 	"github.com/netbirdio/netbird/flow/client"
@@ -25,16 +23,14 @@ import (
 
 // Manager handles netflow tracking and logging
 type Manager struct {
-	mux               sync.Mutex
+	mux            sync.Mutex
-	shutdownWg        sync.WaitGroup
+	shutdownWg     sync.WaitGroup
-	logger            nftypes.FlowLogger
+	logger         nftypes.FlowLogger
-	flowConfig        *nftypes.FlowConfig
+	flowConfig     *nftypes.FlowConfig
-	conntrack         nftypes.ConnTracker
+	conntrack      nftypes.ConnTracker
-	receiverClient    *client.GRPCClient
+	receiverClient *client.GRPCClient
-	eventsWithoutAcks nftypes.Store
+	publicKey      []byte
-	publicKey         []byte
+	cancel         context.CancelFunc
-	cancel            context.CancelFunc
-	retryInterval     time.Duration
 }
 
 // NewManager creates a new netflow manager
@@ -52,11 +48,9 @@ func NewManager(iface nftypes.IFaceMapper, publicKey []byte, statusRecorder *pee
 	}
 
 	return &Manager{
-		logger:            flowLogger,
+		logger:    flowLogger,
-		conntrack:         ct,
+		conntrack: ct,
-		publicKey:         publicKey,
+		publicKey: publicKey,
-		retryInterval:     time.Second,
-		eventsWithoutAcks: store.NewMemoryStore(),
 	}
 }
 
@@ -72,7 +66,6 @@ func (m *Manager) needsNewClient(previous *nftypes.FlowConfig) bool {
 }
 
 // enableFlow starts components for flow tracking
-// must be called under m.mux lock
 func (m *Manager) enableFlow(previous *nftypes.FlowConfig) error {
 	// first make sender ready so events don't pile up
 	if m.needsNewClient(previous) {
@@ -92,7 +85,6 @@ func (m *Manager) enableFlow(previous *nftypes.FlowConfig) error {
 	return nil
 }
 
-// must be called under m.mux lock
 func (m *Manager) resetClient() error {
 	if m.receiverClient != nil {
 		if err := m.receiverClient.Close(); err != nil {
@@ -115,19 +107,14 @@ func (m *Manager) resetClient() error {
 	ctx, cancel := context.WithCancel(context.Background())
 	m.cancel = cancel
 
-	m.shutdownWg.Add(3)
+	m.shutdownWg.Add(2)
-	flowConfigInterval := m.flowConfig.Interval
 	go func() {
 		defer m.shutdownWg.Done()
-		m.receiveACKs(ctx, flowClient, flowConfigInterval)
+		m.receiveACKs(ctx, flowClient)
 	}()
 	go func() {
 		defer m.shutdownWg.Done()
-		m.startSender(ctx, flowConfigInterval)
+		m.startSender(ctx)
-	}()
-	go func() {
-		defer m.shutdownWg.Done()
-		m.startRetries(ctx, flowConfigInterval)
 	}()
 
 	return nil
@@ -211,8 +198,8 @@ func (m *Manager) GetLogger() nftypes.FlowLogger {
 	return m.logger
 }
 
-func (m *Manager) startSender(ctx context.Context, flowConfigInterval time.Duration) {
+func (m *Manager) startSender(ctx context.Context) {
-	ticker := time.NewTicker(flowConfigInterval)
+	ticker := time.NewTicker(m.flowConfig.Interval)
 	defer ticker.Stop()
 
 	for {
@@ -220,29 +207,27 @@ func (m *Manager) startSender(ctx context.Context, flowConfigInterval time.Durat
 		case <-ctx.Done():
 			return
 		case <-ticker.C:
-			collectedEvents := m.logger.ResetAggregationWindow()
+			events := m.logger.GetEvents()
-			events := collectedEvents.GetAggregatedEvents()
 			for _, event := range events {
-				m.eventsWithoutAcks.StoreEvent(event)
 				if err := m.send(event); err != nil {
 					log.Errorf("failed to send flow event to server: %v", err)
-				} else {
+					continue
-					log.Tracef("sent flow event: %s", event.ID)
 				}
+				log.Tracef("sent flow event: %s", event.ID)
 			}
 		}
 	}
 }
 
-func (m *Manager) receiveACKs(ctx context.Context, client *client.GRPCClient, flowConfigInterval time.Duration) {
+func (m *Manager) receiveACKs(ctx context.Context, client *client.GRPCClient) {
-	err := client.Receive(ctx, flowConfigInterval, func(ack *proto.FlowEventAck) error {
+	err := client.Receive(ctx, m.flowConfig.Interval, func(ack *proto.FlowEventAck) error {
 		id, err := uuid.FromBytes(ack.EventId)
 		if err != nil {
 			log.Warnf("failed to convert ack event id to uuid: %v", err)
 			return nil
 		}
 		log.Tracef("received flow event ack: %s", id)
-		m.eventsWithoutAcks.DeleteEvents([]uuid.UUID{id})
+		m.logger.DeleteEvents([]uuid.UUID{id})
 		return nil
 	})
 
@@ -251,51 +236,6 @@ func (m *Manager) receiveACKs(ctx context.Context, client *client.GRPCClient, fl
 	}
 }
 
-// We effectively never drop events (see MaxInterval), which makes eventsWithoutAcks unbounded.
-// We may want to limit the max size of the store, and start dropping oldest events when the threshold is reached.
-func (m *Manager) startRetries(ctx context.Context, flowConfigInterval time.Duration) {
-	timer := time.NewTimer(m.retryInterval)
-	retryBackoff := backoff.WithContext(&backoff.ExponentialBackOff{
-		InitialInterval:     1 * time.Second,
-		RandomizationFactor: 0.5,
-		Multiplier:          1.7,
-		MaxInterval:         flowConfigInterval / 2,
-		MaxElapsedTime:      3 * 30 * 24 * time.Hour, // 3 months
-		Stop:                backoff.Stop,
-		Clock:               backoff.SystemClock,
-	}, ctx)
-	defer timer.Stop()
-
-	for {
-		select {
-		case <-ctx.Done():
-			return
-		case <-timer.C:
-			resetBackoff := true
-			for _, e := range m.eventsWithoutAcks.GetEvents() {
-				if e.Timestamp.Add(time.Second).After(time.Now()) {
-					// grace period on retries to avoid early retries
-					// do not retry if the event is less than 1 sec old
-					continue
-				}
-				if err := m.send(e); err != nil {
-					if nextBackoff := retryBackoff.NextBackOff(); nextBackoff != backoff.Stop {
-						timer = time.NewTimer(nextBackoff)
-						resetBackoff = false
-					} else {
-						resetBackoff = true // we exhausted retries, reset retry loop
-					}
-					break
-				}
-			}
-			if resetBackoff { // use regular retry interval in absence of network errors
-				retryBackoff.Reset()
-				timer = time.NewTimer(m.retryInterval)
-			}
-		}
-	}
-}
-
 func (m *Manager) send(event *nftypes.Event) error {
 	m.mux.Lock()
 	client := m.receiverClient
@@ -310,11 +250,9 @@ func (m *Manager) send(event *nftypes.Event) error {
 
 func toProtoEvent(publicKey []byte, event *nftypes.Event) *proto.FlowEvent {
 	protoEvent := &proto.FlowEvent{
-		EventId:     event.ID[:],
+		EventId:   event.ID[:],
-		Timestamp:   timestamppb.New(event.Timestamp),
+		Timestamp: timestamppb.New(event.Timestamp),
-		PublicKey:   publicKey,
+		PublicKey: publicKey,
-		WindowStart: timestamppb.New(event.WindowStart),
-		WindowEnd:   timestamppb.New(event.WindowEnd),
 		FlowFields: &proto.FlowFields{
 			FlowId:           event.FlowID[:],
 			RuleId:           event.RuleID,
@@ -329,9 +267,6 @@ func toProtoEvent(publicKey []byte, event *nftypes.Event) *proto.FlowEvent {
 			TxBytes:          event.TxBytes,
 			SourceResourceId: event.SourceResourceID,
 			DestResourceId:   event.DestResourceID,
-			NumOfStarts:      event.NumOfStarts,
-			NumOfEnds:        event.NumOfEnds,
-			NumOfDrops:       event.NumOfDrops,
 		},
 	}
 

client/internal/netflow/manager_integration_test.go

@@ -1,291 +0,0 @@
-package netflow
-
-import (
-	"context"
-	"errors"
-	"fmt"
-	"net"
-	"net/netip"
-	"slices"
-	"testing"
-	"time"
-
-	"github.com/google/uuid"
-	"github.com/netbirdio/netbird/client/iface/wgaddr"
-	"github.com/netbirdio/netbird/client/internal/netflow/types"
-	"github.com/netbirdio/netbird/flow/proto"
-	"github.com/stretchr/testify/assert"
-	"github.com/stretchr/testify/require"
-
-	"google.golang.org/grpc"
-)
-
-type testServer struct {
-	proto.UnimplementedFlowServiceServer
-	events         chan *proto.FlowEvent
-	acks           chan *proto.FlowEventAck
-	grpcSrv        *grpc.Server
-	addr           string
-	handlerDone    chan struct{} // signaled each time Events() exits
-	handlerStarted chan struct{} // signaled each time Events() begins
-}
-
-func newTestServer(t *testing.T) *testServer {
-	listener, err := net.Listen("tcp", "127.0.0.1:0")
-	require.NoError(t, err)
-
-	s := &testServer{
-		events:         make(chan *proto.FlowEvent, 100),
-		acks:           make(chan *proto.FlowEventAck, 100),
-		grpcSrv:        grpc.NewServer(),
-		addr:           listener.Addr().String(),
-		handlerDone:    make(chan struct{}, 10),
-		handlerStarted: make(chan struct{}, 10),
-	}
-
-	proto.RegisterFlowServiceServer(s.grpcSrv, s)
-
-	go func() {
-		if err := s.grpcSrv.Serve(listener); err != nil && !errors.Is(err, grpc.ErrServerStopped) {
-			t.Logf("server error: %v", err)
-		}
-	}()
-
-	t.Cleanup(func() {
-		s.grpcSrv.Stop()
-	})
-
-	return s
-}
-
-func (s *testServer) Events(stream proto.FlowService_EventsServer) error {
-	defer func() {
-		select {
-		case s.handlerDone <- struct{}{}:
-		default:
-		}
-	}()
-
-	err := stream.Send(&proto.FlowEventAck{IsInitiator: true})
-	if err != nil {
-		return err
-	}
-
-	select {
-	case s.handlerStarted <- struct{}{}:
-	default:
-	}
-
-	ctx, cancel := context.WithCancel(stream.Context())
-	defer cancel()
-
-	go func() {
-		defer cancel()
-		for {
-			event, err := stream.Recv()
-			if err != nil {
-				return
-			}
-
-			if !event.IsInitiator {
-				select {
-				case s.events <- event:
-				case <-ctx.Done():
-					return
-				}
-			}
-		}
-	}()
-
-	for {
-		select {
-		case ack := <-s.acks:
-			if err := stream.Send(ack); err != nil {
-				return err
-			}
-		case <-ctx.Done():
-			return ctx.Err()
-		}
-	}
-}
-
-func TestSendEventReceiveAck(t *testing.T) {
-	_, cancel := context.WithTimeout(context.Background(), 10*time.Second)
-	t.Cleanup(cancel)
-
-	server := newTestServer(t)
-	manager := createManager(t, server.addr, 60*time.Second) // set high to prevent retries in this test
-	defer manager.Close()
-
-	assert.Eventually(t, func() bool {
-		select {
-		case <-server.handlerStarted:
-			return true
-		default:
-			return false
-		}
-	}, 3*time.Second, 100*time.Millisecond)
-
-	event1 := types.EventFields{
-		FlowID:    uuid.New(),
-		Type:      types.TypeStart,
-		Direction: types.Ingress,
-		DestIP:    ipAddr("172.16.1.2"),
-		DestPort:  2345,
-		Protocol:  6,
-	}
-	manager.logger.StoreEvent(event1)
-	event2 := types.EventFields{
-		FlowID:    uuid.New(),
-		Type:      types.TypeStart,
-		Direction: types.Ingress,
-		DestIP:    ipAddr("172.16.1.1"),
-		DestPort:  1234,
-		Protocol:  6,
-	}
-	manager.logger.StoreEvent(event2)
-
-	// verify the server received logged events
-	serverSideEvents := make([]*proto.FlowEvent, 0)
-	assert.Eventually(t, func() bool {
-		select {
-		case event := <-server.events:
-			serverSideEvents = append(serverSideEvents, event)
-			if len(serverSideEvents) == 2 {
-				return true
-			}
-		default:
-			if len(serverSideEvents) == 2 {
-				return true
-			}
-		}
-		return false
-	}, 5*time.Second, 100*time.Millisecond)
-
-	serverSideFlowIds := make([]uuid.UUID, 0, 2)
-	slices.Values(serverSideEvents)(func(e *proto.FlowEvent) bool {
-		id, err := uuid.FromBytes(e.FlowFields.FlowId)
-		assert.NoError(t, err)
-		serverSideFlowIds = append(serverSideFlowIds, id)
-		return true
-	})
-	assert.ElementsMatch(t, []uuid.UUID{event1.FlowID, event2.FlowID}, serverSideFlowIds)
-
-	// verify the manager tracks un-acked events
-	unackedEvents := manager.eventsWithoutAcks.GetEvents()
-	assert.Len(t, unackedEvents, 2)
-	flowIds := make([]uuid.UUID, 0)
-	slices.Values(unackedEvents)(func(e *types.Event) bool {
-		flowIds = append(flowIds, e.FlowID)
-		return true
-	})
-	assert.ElementsMatch(t, flowIds, []uuid.UUID{event1.FlowID, event2.FlowID})
-}
-
-// verify handling of retries:
-//   - unacked events are retried
-//   - when acks arrive, events are removed from the un-acked event tracker
-func TestRetryEvents(t *testing.T) {
-	_, cancel := context.WithTimeout(context.Background(), 10*time.Second)
-	t.Cleanup(cancel)
-
-	server := newTestServer(t)
-	manager := createManager(t, server.addr, time.Second) // set low to start retries sooner
-	defer manager.Close()
-
-	assert.Eventually(t, func() bool {
-		select {
-		case <-server.handlerStarted:
-			return true
-		default:
-			return false
-		}
-	}, 3*time.Second, 100*time.Millisecond)
-
-	event1 := types.EventFields{
-		FlowID:    uuid.New(),
-		Type:      types.TypeStart,
-		Direction: types.Ingress,
-		DestIP:    ipAddr("172.16.1.2"),
-		DestPort:  2345,
-		Protocol:  6,
-	}
-	manager.logger.StoreEvent(event1)
-	event2 := types.EventFields{
-		FlowID:    uuid.New(),
-		Type:      types.TypeStart,
-		Direction: types.Ingress,
-		DestIP:    ipAddr("172.16.1.1"),
-		DestPort:  1234,
-		Protocol:  6,
-	}
-	manager.logger.StoreEvent(event2)
-
-	// verify the server received retries of logged events
-	serverSideEvents := make([]*proto.FlowEvent, 0)
-	func() {
-		c := time.After(2500 * time.Millisecond)
-		for {
-			select {
-			case event := <-server.events:
-				serverSideEvents = append(serverSideEvents, event)
-			case <-c:
-				return
-			}
-		}
-	}()
-	assert.True(t, len(serverSideEvents) > 2) // must see retries
-
-	uniqueServerSideEvents := make(map[uuid.UUID]*proto.FlowEvent)
-	slices.Values(serverSideEvents)(func(e *proto.FlowEvent) bool {
-		id, err := uuid.FromBytes(e.FlowFields.FlowId)
-		assert.NoError(t, err)
-		uniqueServerSideEvents[id] = e
-		return true
-	})
-	assert.Contains(t, uniqueServerSideEvents, event1.FlowID)
-	assert.Contains(t, uniqueServerSideEvents, event2.FlowID)
-
-	// ack events
-	server.acks <- &proto.FlowEventAck{EventId: uniqueServerSideEvents[event1.FlowID].EventId}
-	server.acks <- &proto.FlowEventAck{EventId: uniqueServerSideEvents[event2.FlowID].EventId}
-
-	assert.EventuallyWithT(t, func(c *assert.CollectT) {
-		unackedEvents := manager.eventsWithoutAcks.GetEvents()
-		assert.Empty(c, unackedEvents)
-
-	}, 3*time.Second, 100*time.Millisecond)
-}
-
-func createManager(t *testing.T, serverAddr string, retryInterval time.Duration) *Manager {
-	t.Helper()
-
-	mockIFace := &mockIFaceMapper{
-		address: wgaddr.Address{
-			Network: netip.MustParsePrefix("192.168.1.1/32"),
-		},
-		isUserspaceBind: true,
-	}
-
-	publicKey := []byte("test-public-key")
-	manager := NewManager(mockIFace, publicKey, nil)
-	manager.retryInterval = retryInterval
-
-	initialConfig := &types.FlowConfig{
-		Enabled:        true,
-		URL:            fmt.Sprintf("http://%s", serverAddr),
-		TokenPayload:   "initial-payload",
-		TokenSignature: "initial-signature",
-		Interval:       500 * time.Millisecond,
-	}
-
-	err := manager.Update(initialConfig)
-	require.NoError(t, err)
-
-	return manager
-}
-
-func ipAddr(a string) netip.Addr {
-	addr, _ := netip.ParseAddr(a)
-	return addr
-}

client/internal/netflow/store/event_aggregation_test.go

@@ -1,365 +0,0 @@
-package store
-
-import (
-	"math/rand"
-	"net/netip"
-	"testing"
-	"time"
-
-	"github.com/google/uuid"
-	"github.com/netbirdio/netbird/client/internal/netflow/types"
-	"github.com/stretchr/testify/assert"
-)
-
-var random = rand.New(rand.NewSource(time.Now().UnixNano()))
-
-func TestFlowAggregation(t *testing.T) {
-	var protocols = []types.Protocol{types.ICMP, types.ICMPv6, types.TCP, types.UDP}
-	var tests = []struct {
-		description string
-		addresses   [][]netip.Addr
-		dstPort     uint16
-		eventTypes  []types.Type
-	}{
-		{
-			description: "start and stop",
-			addresses:   [][]netip.Addr{{netip.MustParseAddr("1.1.1.1"), netip.MustParseAddr("2.2.2.2")}, {netip.MustParseAddr("3.3.3.3"), netip.MustParseAddr("2.2.2.2")}},
-			dstPort:     uint16(random.Uint32() >> 16),
-			eventTypes:  []types.Type{types.TypeStart, types.TypeEnd},
-		},
-		{
-			description: "start and drop",
-			addresses:   [][]netip.Addr{{netip.MustParseAddr("1.1.1.1"), netip.MustParseAddr("2.2.2.2")}, {netip.MustParseAddr("3.3.3.3"), netip.MustParseAddr("2.2.2.2")}},
-			dstPort:     uint16(random.Uint32() >> 16),
-			eventTypes:  []types.Type{types.TypeStart, types.TypeDrop},
-		},
-		{
-			description: "start only",
-			addresses:   [][]netip.Addr{{netip.MustParseAddr("1.1.1.1"), netip.MustParseAddr("2.2.2.2")}, {netip.MustParseAddr("3.3.3.3"), netip.MustParseAddr("2.2.2.2")}},
-			dstPort:     uint16(random.Uint32() >> 16),
-			eventTypes:  []types.Type{types.TypeStart},
-		},
-		{
-			description: "drop only",
-			addresses:   [][]netip.Addr{{netip.MustParseAddr("1.1.1.1"), netip.MustParseAddr("2.2.2.2")}, {netip.MustParseAddr("3.3.3.3"), netip.MustParseAddr("2.2.2.2")}},
-			dstPort:     uint16(random.Uint32() >> 16),
-			eventTypes:  []types.Type{types.TypeDrop},
-		}}
-
-	for _, protocol := range protocols {
-		for _, tt := range tests {
-			t.Run(tt.description+" "+protocol.String(), func(t *testing.T) {
-				store := NewAggregatingMemoryStore()
-				store.WindowEnd = time.Now().Add(5 * time.Second)
-
-				allExpected := make([]*types.Event, 0)
-
-				for _, srcAndDst := range tt.addresses {
-					inEvents, expected := generateEvents(srcAndDst[0], srcAndDst[1], tt.dstPort, tt.eventTypes, protocol, types.Ingress, 0, store.WindowStart, store.WindowEnd)
-					for _, e := range inEvents {
-						store.StoreEvent(e)
-					}
-					allExpected = append(allExpected, expected)
-				}
-
-				events := store.GetAggregatedEvents()
-				assert.ElementsMatch(t, events, allExpected)
-			})
-		}
-	}
-}
-
-func TestIcmpEventAggregation(t *testing.T) {
-	var protocols = []types.Protocol{types.ICMP, types.ICMPv6}
-	var icmpTypes = []uint8{1, 2, 3}
-
-	var tests = []struct {
-		description string
-		addresses   [][]netip.Addr
-		eventTypes  []types.Type
-	}{
-		{
-			description: "start and stop",
-			addresses:   [][]netip.Addr{{netip.MustParseAddr("1.1.1.1"), netip.MustParseAddr("2.2.2.2")}},
-			eventTypes:  []types.Type{types.TypeStart, types.TypeEnd},
-		},
-		{
-			description: "start and drop",
-			addresses:   [][]netip.Addr{{netip.MustParseAddr("1.1.1.1"), netip.MustParseAddr("2.2.2.2")}},
-			eventTypes:  []types.Type{types.TypeStart, types.TypeDrop},
-		},
-		{
-			description: "start only",
-			addresses:   [][]netip.Addr{{netip.MustParseAddr("1.1.1.1"), netip.MustParseAddr("2.2.2.2")}},
-			eventTypes:  []types.Type{types.TypeStart},
-		},
-		{
-			description: "drop only",
-			addresses:   [][]netip.Addr{{netip.MustParseAddr("1.1.1.1"), netip.MustParseAddr("2.2.2.2")}},
-			eventTypes:  []types.Type{types.TypeDrop},
-		}}
-
-	for _, protocol := range protocols {
-		for _, tt := range tests {
-			t.Run(tt.description+" "+protocol.String(), func(t *testing.T) {
-				store := NewAggregatingMemoryStore()
-				store.WindowEnd = time.Now().Add(5 * time.Second)
-
-				allExpected := make([]*types.Event, 0)
-				for _, icmpType := range icmpTypes {
-					events, expected := generateEvents(tt.addresses[0][0], tt.addresses[0][1], 0, tt.eventTypes, protocol, types.Ingress, icmpType, store.WindowStart, store.WindowEnd)
-					for _, e := range events {
-						store.StoreEvent(e)
-					}
-					allExpected = append(allExpected, expected)
-				}
-				aggregatedEvents := store.GetAggregatedEvents()
-				assert.Len(t, aggregatedEvents, len(allExpected))
-				assert.ElementsMatch(t, aggregatedEvents, allExpected)
-			})
-		}
-	}
-}
-
-func TestFlowAggregationOfUnknownProtocols(t *testing.T) {
-	var tests = []struct {
-		description string
-		addresses   [][]netip.Addr
-		dstPort     uint16
-		eventTypes  []types.Type
-	}{
-		{
-			description: "start and stop",
-			addresses:   [][]netip.Addr{{netip.MustParseAddr("1.1.1.1"), netip.MustParseAddr("2.2.2.2")}, {netip.MustParseAddr("3.3.3.3"), netip.MustParseAddr("2.2.2.2")}},
-			dstPort:     uint16(random.Uint32() >> 16),
-			eventTypes:  []types.Type{types.TypeStart, types.TypeEnd},
-		},
-		{
-			description: "start and drop",
-			addresses:   [][]netip.Addr{{netip.MustParseAddr("1.1.1.1"), netip.MustParseAddr("2.2.2.2")}, {netip.MustParseAddr("3.3.3.3"), netip.MustParseAddr("2.2.2.2")}},
-			dstPort:     uint16(random.Uint32() >> 16),
-			eventTypes:  []types.Type{types.TypeStart, types.TypeDrop},
-		},
-		{
-			description: "start only",
-			addresses:   [][]netip.Addr{{netip.MustParseAddr("1.1.1.1"), netip.MustParseAddr("2.2.2.2")}, {netip.MustParseAddr("3.3.3.3"), netip.MustParseAddr("2.2.2.2")}},
-			dstPort:     uint16(random.Uint32() >> 16),
-			eventTypes:  []types.Type{types.TypeStart},
-		},
-		{
-			description: "drop only",
-			addresses:   [][]netip.Addr{{netip.MustParseAddr("1.1.1.1"), netip.MustParseAddr("2.2.2.2")}, {netip.MustParseAddr("3.3.3.3"), netip.MustParseAddr("2.2.2.2")}},
-			dstPort:     uint16(random.Uint32() >> 16),
-			eventTypes:  []types.Type{types.TypeDrop},
-		}}
-
-	for _, tt := range tests {
-		t.Run(tt.description+" "+types.ProtocolUnknown.String(), func(t *testing.T) {
-			store := NewAggregatingMemoryStore()
-			store.WindowEnd = time.Now().Add(5 * time.Second)
-
-			allExpected := make([]*types.Event, 0)
-
-			for _, srcAndDst := range tt.addresses {
-				inEvents, expected := generateEventsForUnknownProtocol(srcAndDst[0], srcAndDst[1], tt.dstPort, tt.eventTypes, types.ProtocolUnknown, types.Ingress, store.WindowStart, store.WindowEnd)
-				for _, e := range inEvents {
-					store.StoreEvent(e)
-				}
-				allExpected = append(allExpected, expected...)
-			}
-
-			events := store.GetAggregatedEvents()
-			assert.ElementsMatch(t, events, allExpected)
-		})
-	}
-}
-
-func TestResetAggregationWindow(t *testing.T) {
-	now := time.Now()
-	nowFunc := func() time.Time { return now }
-	store := NewAggregatingMemoryStoreWithTimeFunc(nowFunc)
-	store.StoreEvent(&types.Event{
-		ID:        uuid.New(),
-		Timestamp: time.Now(),
-		EventFields: types.EventFields{
-			FlowID:           uuid.New(),
-			Type:             types.TypeStart,
-			Protocol:         types.TCP,
-			RuleID:           []byte("rule-id-1"),
-			Direction:        types.Ingress,
-			SourceIP:         netip.MustParseAddr("1.1.1.1"),
-			SourcePort:       1234,
-			DestIP:           netip.MustParseAddr("2.2.2.2"),
-			DestPort:         5678,
-			SourceResourceID: []byte("source-resource-id"),
-			DestResourceID:   []byte("dest-resource-id"),
-			RxPackets:        random.Uint64(),
-			TxPackets:        random.Uint64(),
-			RxBytes:          random.Uint64(),
-			TxBytes:          random.Uint64(),
-		},
-	})
-
-	now = now.Add(1 * time.Second)
-	reset := store.ResetAggregationWindow()
-	previousEvents, ok := reset.(*AggregatingMemory)
-	assert.True(t, ok)
-	assert.NotEqual(t, previousEvents.WindowStart, store.WindowStart)
-	assert.Equal(t, previousEvents.WindowEnd, store.WindowStart)
-	assert.NotEmpty(t, previousEvents.events)
-	assert.Empty(t, store.events)
-}
-
-func generateEvents(srcIp, dstIp netip.Addr, dstPort uint16, eventTypes []types.Type, protocol types.Protocol,
-	direction types.Direction, icmpType uint8, windowStart, windowEnd time.Time) ([]*types.Event, *types.Event) {
-	var rxPackets, txPackets, rxBytes, txBytes uint64
-	inEvents := make([]*types.Event, 0)
-	ts := time.Now()
-	flowId := uuid.New()
-	srcPort := uint16(random.Uint32() >> 16)
-
-	for idx, eventType := range eventTypes {
-		e := &types.Event{
-			ID:        uuid.New(),
-			Timestamp: ts.Add(time.Duration(idx) * time.Second),
-			EventFields: types.EventFields{
-				FlowID:           flowId,
-				Type:             eventType,
-				Protocol:         protocol,
-				RuleID:           []byte("rule-id-1"),
-				Direction:        direction,
-				SourceIP:         srcIp,
-				SourcePort:       srcPort,
-				DestIP:           dstIp,
-				DestPort:         dstPort,
-				SourceResourceID: []byte("source-resource-id"),
-				DestResourceID:   []byte("dest-resource-id"),
-				RxPackets:        random.Uint64(),
-				TxPackets:        random.Uint64(),
-				RxBytes:          random.Uint64(),
-				TxBytes:          random.Uint64(),
-			}}
-		rxBytes += e.RxBytes
-		txBytes += e.TxBytes
-		rxPackets += e.RxPackets
-		txPackets += e.TxPackets
-		inEvents = append(inEvents, e)
-		if protocol == types.ICMP || protocol == types.ICMPv6 {
-			e.ICMPType = icmpType
-		}
-	}
-
-	var start, end, drop uint64
-	for _, eventType := range eventTypes {
-		switch eventType {
-		case types.TypeStart:
-			start += 1
-		case types.TypeDrop:
-			drop += 1
-		case types.TypeEnd:
-			end += 1
-		}
-	}
-	aggregatedEvent := &types.Event{
-		ID:          inEvents[0].ID,
-		Timestamp:   inEvents[0].Timestamp,
-		WindowStart: windowStart,
-		WindowEnd:   windowEnd,
-		EventFields: types.EventFields{
-			FlowID:           flowId,
-			Type:             types.TypeUnknown,
-			Protocol:         inEvents[0].Protocol,
-			RuleID:           []byte("rule-id-1"),
-			Direction:        inEvents[0].Direction,
-			SourceIP:         srcIp,
-			SourcePort:       srcPort,
-			DestIP:           dstIp,
-			DestPort:         dstPort,
-			SourceResourceID: []byte("source-resource-id"),
-			DestResourceID:   []byte("dest-resource-id"),
-			RxPackets:        rxPackets,
-			TxPackets:        txPackets,
-			RxBytes:          rxBytes,
-			TxBytes:          txBytes,
-			NumOfStarts:      start,
-			NumOfEnds:        end,
-			NumOfDrops:       drop,
-		}}
-	if protocol == types.ICMP || protocol == types.ICMPv6 {
-		aggregatedEvent.ICMPType = icmpType
-	}
-
-	return inEvents, aggregatedEvent
-}
-
-func generateEventsForUnknownProtocol(srcIp, dstIp netip.Addr, dstPort uint16, eventTypes []types.Type, protocol types.Protocol,
-	direction types.Direction, windowStart, windowEnd time.Time) ([]*types.Event, []*types.Event) {
-	inEvents := make([]*types.Event, 0)
-	expectedEvents := make([]*types.Event, 0)
-
-	ts := time.Now()
-	flowId := uuid.New()
-	srcPort := uint16(random.Uint32() >> 16)
-
-	for idx, eventType := range eventTypes {
-		e := &types.Event{
-			ID:        uuid.New(),
-			Timestamp: ts.Add(time.Duration(idx) * time.Second),
-			EventFields: types.EventFields{
-				FlowID:           flowId,
-				Type:             eventType,
-				Protocol:         protocol,
-				RuleID:           []byte("rule-id-1"),
-				Direction:        direction,
-				SourceIP:         srcIp,
-				SourcePort:       srcPort,
-				DestIP:           dstIp,
-				DestPort:         dstPort,
-				SourceResourceID: []byte("source-resource-id"),
-				DestResourceID:   []byte("dest-resource-id"),
-				RxPackets:        random.Uint64(),
-				TxPackets:        random.Uint64(),
-				RxBytes:          random.Uint64(),
-				TxBytes:          random.Uint64(),
-			}}
-		inEvents = append(inEvents, e)
-
-		var start, end, drop uint64
-		switch eventType {
-		case types.TypeStart:
-			start = 1
-		case types.TypeDrop:
-			drop = 1
-		case types.TypeEnd:
-			end = 1
-		}
-
-		expectedEvents = append(expectedEvents, &types.Event{
-			ID:          e.ID,
-			Timestamp:   e.Timestamp,
-			WindowStart: windowStart,
-			WindowEnd:   windowEnd,
-			EventFields: types.EventFields{
-				FlowID:           flowId,
-				Type:             types.TypeUnknown,
-				Protocol:         e.Protocol,
-				RuleID:           []byte("rule-id-1"),
-				Direction:        e.Direction,
-				SourceIP:         srcIp,
-				SourcePort:       srcPort,
-				DestIP:           dstIp,
-				DestPort:         dstPort,
-				SourceResourceID: []byte("source-resource-id"),
-				DestResourceID:   []byte("dest-resource-id"),
-				RxPackets:        e.RxPackets,
-				TxPackets:        e.TxPackets,
-				RxBytes:          e.RxBytes,
-				TxBytes:          e.TxBytes,
-				NumOfStarts:      start,
-				NumOfEnds:        end,
-				NumOfDrops:       drop,
-			}})
-	}
-
-	return inEvents, expectedEvents
-}

client/internal/netflow/store/memory.go

@@ -1,15 +1,10 @@
 package store
 
 import (
-	"maps"
-	"math/rand"
-	v2 "math/rand/v2"
-	"net/netip"
-	"slices"
 	"sync"
-	"time"
 
 	"github.com/google/uuid"
+
 	"github.com/netbirdio/netbird/client/internal/netflow/types"
 )
 
@@ -24,14 +19,6 @@ type Memory struct {
 	events map[uuid.UUID]*types.Event
 }
 
-type AggregatingMemory struct {
-	Memory
-	WindowStart time.Time
-	WindowEnd   time.Time
-	rnd         *v2.PCG
-	nowFunc     func() time.Time
-}
-
 func (m *Memory) StoreEvent(event *types.Event) {
 	m.mux.Lock()
 	defer m.mux.Unlock()
@@ -61,104 +48,3 @@ func (m *Memory) DeleteEvents(ids []uuid.UUID) {
 		delete(m.events, id)
 	}
 }
-
-func NewAggregatingMemoryStore() *AggregatingMemory {
-	return NewAggregatingMemoryStoreWithTimeFunc(defaultNowFunc)
-}
-
-// used in tests when deterministic (less random) time intervals are required
-func NewAggregatingMemoryStoreWithTimeFunc(nowFunc func() time.Time) *AggregatingMemory {
-	return &AggregatingMemory{WindowStart: nowFunc(), Memory: Memory{events: make(map[uuid.UUID]*types.Event)}, nowFunc: nowFunc, rnd: v2.NewPCG(rand.Uint64(), rand.Uint64())}
-}
-
-func (am *AggregatingMemory) ResetAggregationWindow() types.FlowEventAggregator {
-	am.mux.Lock()
-	defer am.mux.Unlock()
-
-	now := am.nowFunc()
-	toret := AggregatingMemory{WindowStart: am.WindowStart, WindowEnd: now, Memory: Memory{events: am.events}, rnd: v2.NewPCG(rand.Uint64(), rand.Uint64())}
-
-	am.events = make(map[uuid.UUID]*types.Event)
-	am.WindowStart = now
-
-	return &toret
-}
-
-type aggregationKey struct {
-	srcAddr   netip.Addr
-	destAddr  netip.Addr
-	destPort  uint16
-	direction int
-	protocol  uint8
-	icmpType  uint8
-	unique    uint64 // used to prevent aggregation on non icmp/udp/tcp events
-}
-
-func (am *AggregatingMemory) GetAggregatedEvents() []*types.Event {
-	am.mux.Lock()
-	defer am.mux.Unlock()
-
-	aggregated := make(map[aggregationKey]*types.Event)
-	for _, v := range am.events {
-		lookupKey := aggregationKey{srcAddr: v.SourceIP, destAddr: v.DestIP, destPort: v.DestPort, direction: int(v.Direction), protocol: uint8(v.Protocol), icmpType: v.ICMPType}
-		if _, ok := aggregated[lookupKey]; !ok {
-			event := v.Clone()
-
-			switch event.Type {
-			case types.TypeStart:
-				event.NumOfStarts += 1
-			case types.TypeDrop:
-				event.NumOfDrops += 1
-			case types.TypeEnd:
-				event.NumOfEnds += 1
-			}
-			event.Type = types.TypeUnknown
-
-			// Please note that ICMPCode field isn't propagated by the manager (see flow/proto/flow.pb.go, FlowFields struct)
-			// so the field value in an icmp event in the "aggregated" doesn't matter
-
-			event.WindowStart = am.WindowStart
-			event.WindowEnd = am.WindowEnd
-
-			if event.Protocol != types.ICMP && event.Protocol != types.ICMPv6 && event.Protocol != types.UDP && event.Protocol != types.TCP {
-				lookupKey.unique = am.rnd.Uint64() // to make the lookup key unique so we don't aggregate on it
-			}
-
-			aggregated[lookupKey] = event
-			continue
-		}
-
-		aggregatedEvent := aggregated[lookupKey]
-		if aggregatedEvent.Protocol != types.ICMP && aggregatedEvent.Protocol != types.ICMPv6 && aggregatedEvent.Protocol != types.UDP && aggregatedEvent.Protocol != types.TCP {
-			continue // we don't aggregate this type of events; shouldn't ever get here
-		}
-
-		// track the number of connections, duration?, open and close events?
-		aggregatedEvent.RxBytes += v.RxBytes
-		aggregatedEvent.RxPackets += v.RxPackets
-		aggregatedEvent.TxBytes += v.TxBytes
-		aggregatedEvent.TxPackets += v.TxPackets
-		switch v.Type {
-		case types.TypeStart:
-			aggregatedEvent.NumOfStarts += 1
-		case types.TypeDrop:
-			aggregatedEvent.NumOfDrops += 1
-		case types.TypeEnd:
-			aggregatedEvent.NumOfEnds += 1
-		}
-		if aggregatedEvent.Timestamp.Compare(v.Timestamp) > 0 {
-			aggregatedEvent.Timestamp = v.Timestamp
-			aggregatedEvent.ID = v.ID
-			aggregatedEvent.SourcePort = v.SourcePort
-		}
-		if len(aggregatedEvent.RuleID) == 0 && len(v.RuleID) != 0 {
-			aggregatedEvent.RuleID = slices.Clone(v.RuleID)
-		}
-	}
-
-	return slices.Collect(maps.Values(aggregated)) // could return an iterator instead here
-}
-
-func defaultNowFunc() time.Time {
-	return time.Now()
-}

client/internal/netflow/types/types.go

@@ -2,7 +2,6 @@ package types
 
 import (
 	"net/netip"
-	"slices"
 	"strconv"
 	"time"
 
@@ -70,10 +69,8 @@ const (
 )
 
 type Event struct {
-	ID          uuid.UUID
+	ID        uuid.UUID
-	Timestamp   time.Time
+	Timestamp time.Time
-	WindowStart time.Time
-	WindowEnd   time.Time
 	EventFields
 }
 
@@ -95,17 +92,6 @@ type EventFields struct {
 	TxPackets        uint64
 	RxBytes          uint64
 	TxBytes          uint64
-	NumOfStarts      uint64
-	NumOfEnds        uint64
-	NumOfDrops       uint64
-}
-
-func (e *Event) Clone() *Event {
-	toret := *e
-	toret.RuleID = slices.Clone(e.RuleID)
-	toret.SourceResourceID = slices.Clone(e.SourceResourceID)
-	toret.DestResourceID = slices.Clone(e.DestResourceID)
-	return &toret
 }
 
 type FlowConfig struct {
@@ -128,15 +114,13 @@ type FlowManager interface {
 	GetLogger() FlowLogger
 }
 
-type FlowEventAggregator interface {
-	ResetAggregationWindow() FlowEventAggregator
-	GetAggregatedEvents() []*Event
-}
-
 type FlowLogger interface {
-	ResetAggregationWindow() FlowEventAggregator
 	// StoreEvent stores a flow event
 	StoreEvent(flowEvent EventFields)
+	// GetEvents returns all stored events
+	GetEvents() []*Event
+	// DeleteEvents deletes events from the store
+	DeleteEvents([]uuid.UUID)
 	// Close closes the logger
 	Close()
 	// Enable enables the flow logger receiver
@@ -156,11 +140,6 @@ type Store interface {
 	Close()
 }
 
-type AggregatingStore interface {
-	FlowEventAggregator
-	Store
-}
-
 // ConnTracker defines the interface for connection tracking functionality
 type ConnTracker interface {
 	// Start begins tracking connections by listening for conntrack events.

client/internal/peer/conn.go

@@ -30,11 +30,6 @@ import (
 	relayClient "github.com/netbirdio/netbird/shared/relay/client"
 )
 
-// wgTimeoutEscalationThreshold is the number of consecutive WireGuard
-// handshake timeouts after which the rosenpass state for the peer is
-// considered desynced and gets reset.
-const wgTimeoutEscalationThreshold = 3
-
 // MetricsRecorder is an interface for recording peer connection metrics
 type MetricsRecorder interface {
 	RecordConnectionStages(
@@ -123,9 +118,6 @@ type Conn struct {
 	wgWatcher       *WGWatcher
 	wgWatcherWg     sync.WaitGroup
 	wgWatcherCancel context.CancelFunc
-	// wgTimeouts counts consecutive WireGuard handshake timeouts without a
-	// successful handshake in between. Guarded by mu.
-	wgTimeouts int
 
 	// used to store the remote Rosenpass key for Relayed connection in case of connection update from ice
 	rosenpassRemoteKey []byte
@@ -203,6 +195,7 @@ func NewConn(config ConnConfig, services ServiceDependencies) (*Conn, error) {
 		statusICE:          worker.NewAtomicStatus(),
 		dumpState:          dumpState,
 		endpointUpdater:    NewEndpointUpdater(connLog, config.WgConfig, isController(config)),
+		wgWatcher:          NewWGWatcher(connLog, config.WgConfig.WgInterface, config.Key, dumpState),
 		metricsRecorder:    services.MetricsRecorder,
 	}
 
@@ -670,12 +663,11 @@ func (conn *Conn) onGuardEvent() {
 	}
 }
 
-func (conn *Conn) onWGDisconnected(watcherCtx context.Context) {
+func (conn *Conn) onWGDisconnected() {
 	conn.mu.Lock()
 	defer conn.mu.Unlock()
 
-	// watcherCtx guards against a stale watcher tearing down a connection that already superseded it.
+	if conn.ctx.Err() != nil {
-	if conn.ctx.Err() != nil || watcherCtx.Err() != nil {
 		return
 	}
 
@@ -691,29 +683,6 @@ func (conn *Conn) onWGDisconnected(watcherCtx context.Context) {
 	default:
 		conn.Log.Debugf("No active connection to close on WG timeout")
 	}
-
-	conn.escalateWGTimeoutLocked()
-}
-
-// escalateWGTimeoutLocked resets the peer's rosenpass state after repeated
-// handshake timeouts. With rosenpass enabled, persistent timeouts mean the
-// preshared keys have desynced; the renewal exchange runs over the dead
-// tunnel and cannot resync them. Reporting the peer disconnected drops its
-// rosenpass state, so the next connection configuration programs the
-// rendezvous key and the tunnel can bootstrap again. Callers must hold mu.
-func (conn *Conn) escalateWGTimeoutLocked() {
-	if conn.config.RosenpassConfig.PubKey == nil {
-		return
-	}
-
-	conn.wgTimeouts++
-	if conn.wgTimeouts < wgTimeoutEscalationThreshold || conn.onDisconnected == nil {
-		return
-	}
-	conn.wgTimeouts = 0
-
-	conn.Log.Warnf("%d consecutive WireGuard handshake timeouts, resetting rosenpass state for peer", wgTimeoutEscalationThreshold)
-	conn.onDisconnected(conn.config.WgConfig.RemoteKey)
 }
 
 func (conn *Conn) updateRelayStatus(relayServerAddr string, rosenpassPubKey []byte, updateTime time.Time) {
@@ -833,39 +802,25 @@ func (conn *Conn) isConnectedOnAllWay() (status guard.ConnStatus) {
 	})
 }
 
-// enableWgWatcherIfNeeded starts a fresh watcher instance per connection attempt, so its
-// lifecycle stays bound to conn.mu and enable/disable can't race an old goroutine's shutdown.
-// Caller must hold conn.mu.
 func (conn *Conn) enableWgWatcherIfNeeded(enabledTime time.Time) {
-	if conn.wgWatcher != nil {
+	if !conn.wgWatcher.PrepareInitialHandshake() {
 		return
 	}
 
-	watcher := NewWGWatcher(conn.Log, conn.config.WgConfig.WgInterface, conn.config.Key, conn.dumpState)
-	watcher.PrepareInitialHandshake()
-
 	wgWatcherCtx, wgWatcherCancel := context.WithCancel(conn.ctx)
-	conn.wgWatcher = watcher
 	conn.wgWatcherCancel = wgWatcherCancel
-
 	conn.wgWatcherWg.Add(1)
 	go func() {
 		defer conn.wgWatcherWg.Done()
-		onDisconnected := func() { conn.onWGDisconnected(wgWatcherCtx) }
+		conn.wgWatcher.EnableWgWatcher(wgWatcherCtx, enabledTime, conn.onWGDisconnected, conn.onWGHandshakeSuccess)
-		watcher.EnableWgWatcher(wgWatcherCtx, enabledTime, onDisconnected, conn.onWGHandshakeSuccess, conn.onWGCheckSuccess)
 	}()
 }
 
-// disableWgWatcherIfNeeded cancels and drops the watcher once no transport is active. It never
-// waits for the goroutine: the timeout path reentrantly calls back here under conn.mu, so
-// blocking would deadlock. Caller must hold conn.mu.
 func (conn *Conn) disableWgWatcherIfNeeded() {
-	if conn.currentConnPriority != conntype.None || conn.wgWatcher == nil {
+	if conn.currentConnPriority == conntype.None && conn.wgWatcherCancel != nil {
-		return
+		conn.wgWatcherCancel()
+		conn.wgWatcherCancel = nil
 	}
-	conn.wgWatcherCancel()
-	conn.wgWatcher = nil
-	conn.wgWatcherCancel = nil
 }
 
 func (conn *Conn) newProxy(remoteConn net.Conn) (wgproxy.Proxy, error) {
@@ -888,9 +843,7 @@ func (conn *Conn) resetEndpoint() {
 		return
 	}
 	conn.Log.Infof("reset wg endpoint")
-	if conn.wgWatcher != nil {
+	conn.wgWatcher.Reset()
-		conn.wgWatcher.Reset()
-	}
 	if err := conn.endpointUpdater.RemoveEndpointAddress(); err != nil {
 		conn.Log.Warnf("failed to remove endpoint address before update: %v", err)
 	}
@@ -939,15 +892,6 @@ func (conn *Conn) onWGHandshakeSuccess(when time.Time) {
 	conn.recordConnectionMetrics()
 }
 
-// onWGCheckSuccess is called for every watcher check that observed a fresh
-// handshake, including handshakes of connections that were already up when
-// the watcher started.
-func (conn *Conn) onWGCheckSuccess() {
-	conn.mu.Lock()
-	conn.wgTimeouts = 0
-	conn.mu.Unlock()
-}
-
 // recordConnectionMetrics records connection stage timestamps as metrics
 func (conn *Conn) recordConnectionMetrics() {
 	if conn.metricsRecorder == nil {

client/internal/peer/conn_test.go

@@ -7,7 +7,6 @@ import (
 	"testing"
 	"time"
 
-	log "github.com/sirupsen/logrus"
 	"github.com/stretchr/testify/assert"
 
 	"github.com/netbirdio/netbird/client/iface"
@@ -305,84 +304,3 @@ func TestConn_presharedKey_RosenpassManaged(t *testing.T) {
 		t.Fatalf("expected non-nil presharedKey before Rosenpass manages PSK")
 	}
 }
-
-func newWGTimeoutTestConn(rosenpassEnabled bool, disconnected *[]string) *Conn {
-	cfg := ConnConfig{
-		Key:      "LLHf3Ma6z6mdLbriAJbqhX7+nM/B71lgw2+91q3LfhU=",
-		LocalKey: "RRHf3Ma6z6mdLbriAJbqhX7+nM/B71lgw2+91q3LfhU=",
-		WgConfig: WgConfig{RemoteKey: "LLHf3Ma6z6mdLbriAJbqhX7+nM/B71lgw2+91q3LfhU="},
-	}
-	if rosenpassEnabled {
-		cfg.RosenpassConfig = RosenpassConfig{PubKey: []byte("dummykey")}
-	}
-
-	conn := &Conn{
-		ctx:           context.Background(),
-		config:        cfg,
-		Log:           log.WithField("peer", cfg.Key),
-		metricsStages: &MetricsStages{},
-	}
-	conn.SetOnDisconnected(func(remotePeer string) {
-		*disconnected = append(*disconnected, remotePeer)
-	})
-	return conn
-}
-
-// TestConn_onWGDisconnected_EscalatesToRosenpassReset: repeated handshake
-// timeouts with rosenpass enabled mean the preshared keys have desynced. The
-// renewal exchange runs over the dead tunnel and cannot resync them, so after
-// wgTimeoutEscalationThreshold consecutive timeouts the conn must report the
-// peer disconnected, dropping its rosenpass state so the next configuration
-// programs the rendezvous key.
-func TestConn_onWGDisconnected_EscalatesToRosenpassReset(t *testing.T) {
-	var disconnected []string
-	conn := newWGTimeoutTestConn(true, &disconnected)
-
-	for i := 0; i < wgTimeoutEscalationThreshold-1; i++ {
-		conn.onWGDisconnected(conn.ctx)
-	}
-	assert.Empty(t, disconnected, "escalation must not fire below the threshold")
-
-	conn.onWGDisconnected(conn.ctx)
-	assert.Equal(t, []string{conn.config.WgConfig.RemoteKey}, disconnected,
-		"reaching the threshold must report the peer disconnected once")
-
-	for i := 0; i < wgTimeoutEscalationThreshold-1; i++ {
-		conn.onWGDisconnected(conn.ctx)
-	}
-	assert.Len(t, disconnected, 1, "escalation must restart counting after firing")
-
-	conn.onWGDisconnected(conn.ctx)
-	assert.Len(t, disconnected, 2, "continued timeouts must escalate again")
-}
-
-// TestConn_onWGDisconnected_CheckSuccessResetsEscalation: a successful
-// handshake between timeouts means the tunnel recovered; the counter must
-// start over.
-func TestConn_onWGDisconnected_CheckSuccessResetsEscalation(t *testing.T) {
-	var disconnected []string
-	conn := newWGTimeoutTestConn(true, &disconnected)
-
-	for i := 0; i < wgTimeoutEscalationThreshold-1; i++ {
-		conn.onWGDisconnected(conn.ctx)
-	}
-	conn.onWGCheckSuccess()
-
-	for i := 0; i < wgTimeoutEscalationThreshold-1; i++ {
-		conn.onWGDisconnected(conn.ctx)
-	}
-	assert.Empty(t, disconnected, "handshake success must reset the timeout count")
-}
-
-// TestConn_onWGDisconnected_NoEscalationWithoutRosenpass: without rosenpass
-// there is no per-peer key state to reset; repeated timeouts must not report
-// disconnects.
-func TestConn_onWGDisconnected_NoEscalationWithoutRosenpass(t *testing.T) {
-	var disconnected []string
-	conn := newWGTimeoutTestConn(false, &disconnected)
-
-	for i := 0; i < wgTimeoutEscalationThreshold*3; i++ {
-		conn.onWGDisconnected(conn.ctx)
-	}
-	assert.Empty(t, disconnected, "escalation must be limited to rosenpass connections")
-}

client/internal/peer/guard/guard.go

@@ -85,11 +85,7 @@ func (g *Guard) reconnectLoopWithRetry(ctx context.Context, callback func()) {
 	defer g.srWatcher.RemoveListener(srReconnectedChan)
 
 	ticker := g.initialTicker(ctx)
-	defer func() {
+	defer ticker.Stop()
-		// If backoff.Ticker.send is blocked, context.Done will not close the Ticker goroutine.
-		// We have to explicitly call Stop, even if we use backoff.WithContext.
-		ticker.Stop()
-	}()
 
 	tickerChannel := ticker.C
 

client/internal/peer/guard/guard_leak_test.go

@@ -1,92 +0,0 @@
-package guard
-
-import (
-	"context"
-	"runtime"
-	"strings"
-	"sync"
-	"testing"
-	"time"
-
-	log "github.com/sirupsen/logrus"
-
-	"github.com/netbirdio/netbird/client/internal/peer/ice"
-)
-
-func newTestGuard(status connStatusFunc) *Guard {
-	srw := NewSRWatcher(nil, nil, nil, ice.Config{})
-	return NewGuard(log.WithField("test", "guard"), status, 50*time.Millisecond, srw)
-}
-
-// countBackoffTickerGoroutines returns how many goroutines are currently sitting
-// in backoff/v4.(*Ticker).run (a ticker goroutine that has not exited).
-func countBackoffTickerGoroutines() int {
-	buf := make([]byte, 1<<25) // 32MB
-	n := runtime.Stack(buf, true)
-	return strings.Count(string(buf[:n]), "backoff/v4.(*Ticker).run")
-}
-
-// TestGuard_ReconnectTicker_NoGoroutineLeakOnShutdown reproduces a observed
-// leak: after a shutdown burst, ticker run/send goroutines stay parked
-// forever even though every reconnect loop has exited.
-func TestGuard_ReconnectTicker_NoGoroutineLeakOnShutdown(t *testing.T) {
-	before := countBackoffTickerGoroutines()
-
-	const peers = 6000
-	cancels := make([]context.CancelFunc, 0, peers)
-	var wg sync.WaitGroup
-
-	// A status check slower than the tick cadence. This models the real
-	// isConnectedOnAllWay/callback doing work: while the loop is busy in the
-	// handler, the ticker fires the next tick and parks in send(), because
-	// send() never selects on ctx.
-	slowStatus := func() ConnStatus {
-		time.Sleep(70 * time.Millisecond)
-		return ConnStatusConnected
-	}
-
-	for range peers {
-		g := newTestGuard(slowStatus)
-		ctx, cancel := context.WithCancel(context.Background())
-		cancels = append(cancels, cancel)
-		wg.Add(1)
-		go func() {
-			defer wg.Done()
-			g.Start(ctx, func() {})
-		}()
-		// Force the live ticker to be a newReconnectTicker.
-		g.SetRelayedConnDisconnected()
-	}
-
-	// Let the replacement tickers get past their 800ms initial interval, so
-	// many are parked in send() waiting on the (slow) consumer when we tear
-	// everything down.
-	time.Sleep(1500 * time.Millisecond)
-
-	// Shutdown burst: cancel every peer at once, like engine teardown.
-	for _, c := range cancels {
-		c()
-	}
-
-	// Every reconnect loop must return
-	waitCh := make(chan struct{})
-	go func() { wg.Wait(); close(waitCh) }()
-	select {
-	case <-waitCh:
-	case <-time.After(30 * time.Second):
-		t.Fatal("not all reconnect loops returned after ctx cancel")
-	}
-
-	// Give any correctly-stopped ticker goroutines time to unwind.
-	for range 50 {
-		runtime.Gosched()
-		time.Sleep(10 * time.Millisecond)
-	}
-
-	leaked := countBackoffTickerGoroutines() - before
-	t.Logf("backoff Ticker.run goroutines still parked after teardown of %d peers: %d", peers, leaked)
-	if leaked > 0 {
-		t.Errorf("LEAK: %d backoff ticker goroutines parked after all reconnect loops exited "+
-			"(defer ticker.Stop() stops the initial ticker, not the live replacement)", leaked)
-	}
-}

client/internal/peer/status.go

@@ -7,7 +7,6 @@ import (
 	"net/netip"
 	"slices"
 	"sync"
-	"sync/atomic"
 	"time"
 
 	"github.com/google/uuid"
@@ -192,30 +191,23 @@ func (s *StatusChangeSubscription) Events() chan map[string]RouterState {
 // every private-service request) don't contend against each other.
 // Pure read methods take RLock; anything that mutates state takes Lock.
 type Status struct {
-	mux                 sync.RWMutex
+	mux                   sync.RWMutex
-	muxRelays           sync.RWMutex
+	muxRelays             sync.RWMutex
-	peers               map[string]State
+	peers                 map[string]State
-	ipToKey             map[string]string
+	ipToKey               map[string]string
-	changeNotify        map[string]map[string]*StatusChangeSubscription // map[peerID]map[subscriptionID]*StatusChangeSubscription
+	changeNotify          map[string]map[string]*StatusChangeSubscription // map[peerID]map[subscriptionID]*StatusChangeSubscription
-	signalState         bool
+	signalState           bool
-	signalError         error
+	signalError           error
-	managementState     bool
+	managementState       bool
-	managementError     error
+	managementError       error
-	relayStates         []relay.ProbeResult
+	relayStates           []relay.ProbeResult
-	localPeer           LocalPeerState
+	localPeer             LocalPeerState
-	offlinePeers        []State
+	offlinePeers          []State
-	mgmAddress          string
+	mgmAddress            string
-	signalAddress       string
+	signalAddress         string
-	notifier            *notifier
+	notifier              *notifier
-	rosenpassEnabled    bool
+	rosenpassEnabled      bool
-	rosenpassPermissive bool
+	rosenpassPermissive   bool
-	// sessionExpiresAt is the absolute UTC instant at which the peer's SSO
-	// session expires. Zero when the peer is not SSO-tracked or login
-	// expiration is disabled. Populated from management LoginResponse /
-	// SyncResponse and exposed via the daemon's Status / SubscribeStatus RPC
-	// so the UI can show remaining time without itself talking to mgm.
-	sessionExpiresAt time.Time
-
 	nsGroupStates         []NSGroupState
 	resolvedDomainsStates map[domain.Domain]ResolvedDomainInfo
 	lazyConnectionEnabled bool
@@ -231,21 +223,6 @@ type Status struct {
 	eventStreams map[string]chan *proto.SystemEvent
 	eventQueue   *EventQueue
 
-	// stateChangeStreams fan-out connection-state changes (connected /
-	// disconnected / connecting / address change / peers list change) to
-	// every active SubscribeStatus gRPC stream. Each subscriber gets a
-	// buffered chan; the notifier non-blockingly pings them so a slow
-	// consumer can never stall the daemon.
-	stateChangeMux     sync.Mutex
-	stateChangeStreams map[string]chan struct{}
-
-	// networksRevision bumps whenever the routed-networks set or their
-	// selected state changes (driven by the route manager). Surfaced in the
-	// status snapshot so the UI can fingerprint on it and re-fetch
-	// ListNetworks only on a real change. Atomic so the snapshot builder can
-	// read it without taking mux.
-	networksRevision atomic.Uint64
-
 	ingressGwMgr *ingressgw.Manager
 
 	routeIDLookup routeIDLookup
@@ -260,7 +237,6 @@ func NewRecorder(mgmAddress string) *Status {
 		changeNotify:          make(map[string]map[string]*StatusChangeSubscription),
 		eventStreams:          make(map[string]chan *proto.SystemEvent),
 		eventQueue:            NewEventQueue(eventQueueSize),
-		stateChangeStreams:    make(map[string]chan struct{}),
 		offlinePeers:          make([]State, 0),
 		notifier:              newNotifier(),
 		mgmAddress:            mgmAddress,
@@ -425,7 +401,6 @@ func (d *Status) UpdatePeerState(receivedState State) error {
 	if notifyRouter {
 		d.dispatchRouterPeers(receivedState.PubKey, routerSnapshot)
 	}
-	d.notifyStateChange()
 	return nil
 }
 
@@ -451,7 +426,6 @@ func (d *Status) AddPeerStateRoute(peer string, route string, resourceId route.R
 
 	// todo: consider to make sense of this notification or not
 	d.notifier.peerListChanged(numPeers)
-	d.notifyStateChange()
 	return nil
 }
 
@@ -477,7 +451,6 @@ func (d *Status) RemovePeerStateRoute(peer string, route string) error {
 
 	// todo: consider to make sense of this notification or not
 	d.notifier.peerListChanged(numPeers)
-	d.notifyStateChange()
 	return nil
 }
 
@@ -527,7 +500,6 @@ func (d *Status) UpdatePeerICEState(receivedState State) error {
 	if notifyRouter {
 		d.dispatchRouterPeers(receivedState.PubKey, routerSnapshot)
 	}
-	d.notifyStateChange()
 	return nil
 }
 
@@ -564,7 +536,6 @@ func (d *Status) UpdatePeerRelayedState(receivedState State) error {
 	if notifyRouter {
 		d.dispatchRouterPeers(receivedState.PubKey, routerSnapshot)
 	}
-	d.notifyStateChange()
 	return nil
 }
 
@@ -600,7 +571,6 @@ func (d *Status) UpdatePeerRelayedStateToDisconnected(receivedState State) error
 	if notifyRouter {
 		d.dispatchRouterPeers(receivedState.PubKey, routerSnapshot)
 	}
-	d.notifyStateChange()
 	return nil
 }
 
@@ -639,7 +609,6 @@ func (d *Status) UpdatePeerICEStateToDisconnected(receivedState State) error {
 	if notifyRouter {
 		d.dispatchRouterPeers(receivedState.PubKey, routerSnapshot)
 	}
-	d.notifyStateChange()
 	return nil
 }
 
@@ -733,7 +702,6 @@ func (d *Status) FinishPeerListModifications() {
 	for _, rd := range dispatches {
 		d.dispatchRouterPeers(rd.peerID, rd.snapshot)
 	}
-	d.notifyStateChange()
 }
 
 func (d *Status) SubscribeToPeerStateChanges(ctx context.Context, peerID string) *StatusChangeSubscription {
@@ -792,36 +760,6 @@ func (d *Status) UpdateLocalPeerState(localPeerState LocalPeerState) {
 	d.mux.Unlock()
 
 	d.notifier.localAddressChanged(fqdn, ip)
-	d.notifyStateChange()
-}
-
-// SetSessionExpiresAt records the absolute UTC instant at which the peer's
-// SSO session is set to expire. Pass the zero value to clear (e.g. when the
-// management server stops publishing a deadline because login expiration was
-// disabled or the peer is not SSO-tracked). Same-value updates are no-ops;
-// real changes fan out via notifyStateChange so SubscribeStatus consumers
-// pick up the new deadline on their next read.
-func (d *Status) SetSessionExpiresAt(deadline time.Time) {
-	d.mux.Lock()
-	if d.sessionExpiresAt.Equal(deadline) {
-		d.mux.Unlock()
-		return
-	}
-	d.sessionExpiresAt = deadline
-	d.mux.Unlock()
-	d.notifyStateChange()
-}
-
-// GetSessionExpiresAt returns the most recently recorded SSO session deadline,
-// or the zero value when no deadline is tracked. A deadline in the past is
-// returned as-is: it means the session has expired, and consumers (tray row,
-// CLI status) render it as "expired" rather than hiding it — masking it as
-// "none" would blank the UI at the exact moment it should say the session
-// ended.
-func (d *Status) GetSessionExpiresAt() time.Time {
-	d.mux.Lock()
-	defer d.mux.Unlock()
-	return d.sessionExpiresAt
 }
 
 // AddLocalPeerStateRoute adds a route to the local peer state
@@ -890,19 +828,11 @@ func (d *Status) CleanLocalPeerState() {
 	d.mux.Unlock()
 
 	d.notifier.localAddressChanged(fqdn, ip)
-	d.notifyStateChange()
 }
 
 // MarkManagementDisconnected sets ManagementState to disconnected
 func (d *Status) MarkManagementDisconnected(err error) {
 	d.mux.Lock()
-	// Health checks re-mark the same state on every probe; skip the fan-out
-	// when nothing actually changed so we don't flood SubscribeStatus
-	// consumers with identical snapshots.
-	if !d.managementState && errors.Is(d.managementError, err) {
-		d.mux.Unlock()
-		return
-	}
 	d.managementState = false
 	d.managementError = err
 	mgm := d.managementState
@@ -910,16 +840,11 @@ func (d *Status) MarkManagementDisconnected(err error) {
 	d.mux.Unlock()
 
 	d.notifier.updateServerStates(mgm, sig)
-	d.notifyStateChange()
 }
 
 // MarkManagementConnected sets ManagementState to connected
 func (d *Status) MarkManagementConnected() {
 	d.mux.Lock()
-	if d.managementState && d.managementError == nil {
-		d.mux.Unlock()
-		return
-	}
 	d.managementState = true
 	d.managementError = nil
 	mgm := d.managementState
@@ -927,7 +852,6 @@ func (d *Status) MarkManagementConnected() {
 	d.mux.Unlock()
 
 	d.notifier.updateServerStates(mgm, sig)
-	d.notifyStateChange()
 }
 
 // UpdateSignalAddress update the address of the signal server
@@ -961,10 +885,6 @@ func (d *Status) UpdateLazyConnection(enabled bool) {
 // MarkSignalDisconnected sets SignalState to disconnected
 func (d *Status) MarkSignalDisconnected(err error) {
 	d.mux.Lock()
-	if !d.signalState && errors.Is(d.signalError, err) {
-		d.mux.Unlock()
-		return
-	}
 	d.signalState = false
 	d.signalError = err
 	mgm := d.managementState
@@ -972,16 +892,11 @@ func (d *Status) MarkSignalDisconnected(err error) {
 	d.mux.Unlock()
 
 	d.notifier.updateServerStates(mgm, sig)
-	d.notifyStateChange()
 }
 
 // MarkSignalConnected sets SignalState to connected
 func (d *Status) MarkSignalConnected() {
 	d.mux.Lock()
-	if d.signalState && d.signalError == nil {
-		d.mux.Unlock()
-		return
-	}
 	d.signalState = true
 	d.signalError = nil
 	mgm := d.managementState
@@ -989,7 +904,6 @@ func (d *Status) MarkSignalConnected() {
 	d.mux.Unlock()
 
 	d.notifier.updateServerStates(mgm, sig)
-	d.notifyStateChange()
 }
 
 func (d *Status) UpdateRelayStates(relayResults []relay.ProbeResult) {
@@ -1196,19 +1110,16 @@ func (d *Status) GetFullStatus() FullStatus {
 // ClientStart will notify all listeners about the new service state
 func (d *Status) ClientStart() {
 	d.notifier.clientStart()
-	d.notifyStateChange()
 }
 
 // ClientStop will notify all listeners about the new service state
 func (d *Status) ClientStop() {
 	d.notifier.clientStop()
-	d.notifyStateChange()
 }
 
 // ClientTeardown will notify all listeners about the service is under teardown
 func (d *Status) ClientTeardown() {
 	d.notifier.clientTearDown()
-	d.notifyStateChange()
 }
 
 // SetConnectionListener set a listener to the notifier
@@ -1350,79 +1261,6 @@ func (d *Status) GetEventHistory() []*proto.SystemEvent {
 	return d.eventQueue.GetAll()
 }
 
-// SubscribeToStateChanges hands back a channel that receives a tick on
-// every connection-state change (connected / disconnected / connecting /
-// address change / peers-list change). The channel is buffered to one
-// pending tick so a coalesced burst still wakes the consumer exactly
-// once. Pass the returned id to UnsubscribeFromStateChanges to detach.
-func (d *Status) SubscribeToStateChanges() (string, <-chan struct{}) {
-	d.stateChangeMux.Lock()
-	defer d.stateChangeMux.Unlock()
-
-	id := uuid.New().String()
-	ch := make(chan struct{}, 1)
-	d.stateChangeStreams[id] = ch
-	return id, ch
-}
-
-// UnsubscribeFromStateChanges releases a SubscribeToStateChanges channel
-// and closes it so any consumer goroutine selecting on the channel
-// unblocks cleanly.
-func (d *Status) UnsubscribeFromStateChanges(id string) {
-	d.stateChangeMux.Lock()
-	defer d.stateChangeMux.Unlock()
-
-	if ch, ok := d.stateChangeStreams[id]; ok {
-		close(ch)
-		delete(d.stateChangeStreams, id)
-	}
-}
-
-// notifyStateChange wakes every SubscribeToStateChanges subscriber. Drops
-// the tick if a subscriber's buffer is full — by definition the consumer
-// is already going to fetch the latest snapshot, so multiple pending ticks
-// would be redundant.
-func (d *Status) notifyStateChange() {
-	d.stateChangeMux.Lock()
-	defer d.stateChangeMux.Unlock()
-
-	for _, ch := range d.stateChangeStreams {
-		select {
-		case ch <- struct{}{}:
-		default:
-		}
-	}
-}
-
-// NotifyStateChange is the public wake-the-subscribers entry point used by
-// callers that mutate state outside the peer recorder — most importantly
-// the connect-state machine, which writes StatusNeedsLogin into the
-// shared contextState (client/internal/state.go) without touching any
-// recorder field. Without this push the SubscribeStatus stream stays on
-// the previous snapshot until an unrelated peer/management/signal
-// change happens to fire notifyStateChange, leaving the UI's status
-// out of sync with the daemon.
-func (d *Status) NotifyStateChange() {
-	d.notifyStateChange()
-}
-
-// BumpNetworksRevision increments the routed-networks revision and wakes every
-// SubscribeStatus subscriber. The route manager calls it when a network map
-// changes the available routes or when a selection is applied — the peer
-// status itself only records actively-routed (chosen) networks, so without
-// this bump a candidate route appearing/disappearing would never reach the UI.
-func (d *Status) BumpNetworksRevision() {
-	d.networksRevision.Add(1)
-	d.notifyStateChange()
-}
-
-// GetNetworksRevision returns the current routed-networks revision, surfaced in
-// the status snapshot so the UI can detect route/selection changes (see
-// BumpNetworksRevision).
-func (d *Status) GetNetworksRevision() uint64 {
-	return d.networksRevision.Load()
-}
-
 func (d *Status) SetWgIface(wgInterface WGIfaceStatus) {
 	d.mux.Lock()
 	defer d.mux.Unlock()

client/internal/peer/status_test.go

@@ -314,39 +314,3 @@ func TestGetFullStatus(t *testing.T) {
 	assert.Equal(t, signalState, fullStatus.SignalState, "signal status should be equal")
 	assert.ElementsMatch(t, []State{peerState1, peerState2}, fullStatus.Peers, "peers states should match")
 }
-
-// notified reports whether a state-change tick is pending on ch, draining it.
-func notified(ch <-chan struct{}) bool {
-	select {
-	case <-ch:
-		return true
-	default:
-		return false
-	}
-}
-
-func TestMarkServerStateDoesNotNotifyWhenUnchanged(t *testing.T) {
-	status := NewRecorder("https://mgm")
-	_, ch := status.SubscribeToStateChanges()
-
-	// First transition is a real change and must notify.
-	status.MarkManagementConnected()
-	require.True(t, notified(ch), "first connect should notify")
-
-	// Re-marking the same state must not notify again.
-	status.MarkManagementConnected()
-	assert.False(t, notified(ch), "redundant connect should not notify")
-
-	// Same for signal.
-	status.MarkSignalConnected()
-	require.True(t, notified(ch), "first signal connect should notify")
-	status.MarkSignalConnected()
-	assert.False(t, notified(ch), "redundant signal connect should not notify")
-
-	// A genuine change (disconnect with an error) notifies again.
-	err := errors.New("boom")
-	status.MarkManagementDisconnected(err)
-	require.True(t, notified(ch), "disconnect should notify")
-	status.MarkManagementDisconnected(err)
-	assert.False(t, notified(ch), "redundant disconnect should not notify")
-}

client/internal/peer/wg_watcher.go

@@ -3,6 +3,7 @@ package peer
 import (
 	"context"
 	"fmt"
+	"sync"
 	"time"
 
 	log "github.com/sirupsen/logrus"
@@ -23,14 +24,14 @@ type WGInterfaceStater interface {
 	GetStats() (map[string]configurer.WGStats, error)
 }
 
-// WGWatcher is single-shot: one instance per connection attempt, run once, then discarded.
-// Lifecycle is owned by Conn under conn.mu, so it keeps no "enabled" state to go stale.
 type WGWatcher struct {
 	log           *log.Entry
 	wgIfaceStater WGInterfaceStater
 	peerKey       string
 	stateDump     *stateDump
 
+	enabled   bool
+	muEnabled sync.Mutex
 	// initialHandshake is not thread-safe; never call PrepareInitialHandshake and EnableWgWatcher concurrently.
 	initialHandshake time.Time
 
@@ -47,23 +48,36 @@ func NewWGWatcher(log *log.Entry, wgIfaceStater WGInterfaceStater, peerKey strin
 	}
 }
 
-// PrepareInitialHandshake reads the peer's current WireGuard handshake time. It must be
+// PrepareInitialHandshake reserves the watcher and reads the peer's current WireGuard
-// called before the peer is (re)configured on the WireGuard interface, so the captured
+// handshake time. It must be called before the peer is (re)configured on the WireGuard
-// baseline reflects the state prior to this connection attempt instead of racing with
+// interface, so the captured baseline reflects the state prior to this connection attempt
-// that configuration.
+// instead of racing with that configuration. Returns ok=false if the watcher is already
-func (w *WGWatcher) PrepareInitialHandshake() {
+// running, in which case EnableWgWatcher must not be called.
+func (w *WGWatcher) PrepareInitialHandshake() (ok bool) {
+	w.muEnabled.Lock()
+	if w.enabled {
+		w.muEnabled.Unlock()
+		return false
+	}
+
 	w.log.Debugf("enable WireGuard watcher")
+	w.enabled = true
+	w.muEnabled.Unlock()
+
 	handshake, _ := w.wgState()
 	w.initialHandshake = handshake
+	return true
 }
 
 // EnableWgWatcher runs the WireGuard watcher loop using the handshake baseline captured by
 // PrepareInitialHandshake. The watcher runs until ctx is cancelled. Caller is responsible
-// for context lifecycle management. onHandshakeSuccessFn is called only for the first
+// for context lifecycle management.
-// handshake observed by this run, onCheckSuccessFn for every check that observed a fresh
+func (w *WGWatcher) EnableWgWatcher(ctx context.Context, enabledTime time.Time, onDisconnectedFn func(), onHandshakeSuccessFn func(when time.Time)) {
-// handshake, including the first.
+	w.periodicHandshakeCheck(ctx, onDisconnectedFn, onHandshakeSuccessFn, enabledTime, w.initialHandshake)
-func (w *WGWatcher) EnableWgWatcher(ctx context.Context, enabledTime time.Time, onDisconnectedFn func(), onHandshakeSuccessFn func(when time.Time), onCheckSuccessFn func()) {
+
-	w.periodicHandshakeCheck(ctx, onDisconnectedFn, onHandshakeSuccessFn, onCheckSuccessFn, enabledTime, w.initialHandshake)
+	w.muEnabled.Lock()
+	w.enabled = false
+	w.muEnabled.Unlock()
 }
 
 // Reset signals the watcher that the WireGuard peer has been reset and a new
@@ -76,7 +90,7 @@ func (w *WGWatcher) Reset() {
 }
 
 // wgStateCheck help to check the state of the WireGuard handshake and relay connection
-func (w *WGWatcher) periodicHandshakeCheck(ctx context.Context, onDisconnectedFn func(), onHandshakeSuccessFn func(when time.Time), onCheckSuccessFn func(), enabledTime time.Time, initialHandshake time.Time) {
+func (w *WGWatcher) periodicHandshakeCheck(ctx context.Context, onDisconnectedFn func(), onHandshakeSuccessFn func(when time.Time), enabledTime time.Time, initialHandshake time.Time) {
 	w.log.Infof("WireGuard watcher started")
 
 	timer := time.NewTimer(wgHandshakeOvertime)
@@ -89,7 +103,6 @@ func (w *WGWatcher) periodicHandshakeCheck(ctx context.Context, onDisconnectedFn
 		case <-timer.C:
 			handshake, ok := w.handshakeCheck(lastHandshake)
 			if !ok {
-				// early ctx cancel check return
 				if ctx.Err() != nil {
 					return
 				}
@@ -104,10 +117,6 @@ func (w *WGWatcher) periodicHandshakeCheck(ctx context.Context, onDisconnectedFn
 				}
 			}
 
-			if onCheckSuccessFn != nil && ctx.Err() == nil {
-				onCheckSuccessFn()
-			}
-
 			lastHandshake = *handshake
 
 			resetTime := time.Until(handshake.Add(checkPeriod))
@@ -138,9 +147,9 @@ func (w *WGWatcher) handshakeCheck(lastHandshake time.Time) (*time.Time, bool) {
 
 	w.log.Tracef("previous handshake, handshake: %v, %v", lastHandshake, handshake)
 
-	// the current known handshake did not change
+	// the current know handshake did not change
 	if handshake.Equal(lastHandshake) {
-		w.log.Warnf("WireGuard handshake not updated: %v", handshake)
+		w.log.Warnf("WireGuard handshake timed out: %v", handshake)
 		return nil, false
 	}
 

client/internal/peer/wg_watcher_test.go

@@ -7,6 +7,7 @@ import (
 	"time"
 
 	log "github.com/sirupsen/logrus"
+	"github.com/stretchr/testify/require"
 
 	"github.com/netbirdio/netbird/client/iface/configurer"
 )
@@ -23,72 +24,6 @@ func (m *MocWgIface) disconnect() {
 	m.stop = true
 }
 
-type mockHandshakeStats struct {
-	mu        sync.Mutex
-	handshake time.Time
-}
-
-func (m *mockHandshakeStats) GetStats() (map[string]configurer.WGStats, error) {
-	m.mu.Lock()
-	defer m.mu.Unlock()
-	return map[string]configurer.WGStats{"": {LastHandshake: m.handshake}}, nil
-}
-
-func (m *mockHandshakeStats) advance() {
-	m.mu.Lock()
-	defer m.mu.Unlock()
-	m.handshake = time.Now()
-}
-
-// TestWGWatcher_CheckSuccessCallback: onCheckSuccessFn must fire for a fresh
-// handshake even when the watcher started with an existing handshake baseline,
-// the case where onHandshakeSuccessFn stays silent.
-func TestWGWatcher_CheckSuccessCallback(t *testing.T) {
-	// checkPeriod bounds how stale a handshake may be before the watcher treats it
-	// as a suspended-machine timeout. The first check fires after wgHandshakeOvertime,
-	// so keep checkPeriod well above any scheduling jitter to avoid a false timeout
-	// converting the expected success into a disconnect on a loaded runner.
-	checkPeriod = 1 * time.Minute
-	wgHandshakeOvertime = 1 * time.Second
-
-	mlog := log.WithField("peer", "tet")
-	// Use an old baseline so advance() yields a strictly newer handshake even on
-	// platforms with coarse clock resolution (Windows), where two time.Now() calls
-	// microseconds apart can return the same instant and read as a timed-out handshake.
-	stats := &mockHandshakeStats{handshake: time.Now().Add(-time.Hour)}
-	watcher := NewWGWatcher(mlog, stats, "", newStateDump("peer", mlog, &Status{}))
-
-	ctx, cancel := context.WithCancel(context.Background())
-	defer cancel()
-
-	watcher.PrepareInitialHandshake()
-
-	firstHandshake := make(chan struct{}, 1)
-	checkSuccess := make(chan struct{}, 1)
-	go watcher.EnableWgWatcher(ctx, time.Now(), func() {}, func(when time.Time) {
-		firstHandshake <- struct{}{}
-	}, func() {
-		select {
-		case checkSuccess <- struct{}{}:
-		default:
-		}
-	})
-
-	stats.advance()
-
-	select {
-	case <-checkSuccess:
-	case <-time.After(10 * time.Second):
-		t.Errorf("timeout waiting for check success callback")
-	}
-
-	select {
-	case <-firstHandshake:
-		t.Errorf("first-handshake callback must not fire for a non-zero baseline")
-	default:
-	}
-}
-
 func TestWGWatcher_EnableWgWatcher(t *testing.T) {
 	checkPeriod = 5 * time.Second
 	wgHandshakeOvertime = 1 * time.Second
@@ -100,7 +35,8 @@ func TestWGWatcher_EnableWgWatcher(t *testing.T) {
 	ctx, cancel := context.WithCancel(context.Background())
 	defer cancel()
 
-	watcher.PrepareInitialHandshake()
+	ok := watcher.PrepareInitialHandshake()
+	require.True(t, ok, "watcher should not be enabled yet")
 
 	onDisconnected := make(chan struct{}, 1)
 	go watcher.EnableWgWatcher(ctx, time.Now(), func() {
@@ -108,7 +44,7 @@ func TestWGWatcher_EnableWgWatcher(t *testing.T) {
 		onDisconnected <- struct{}{}
 	}, func(when time.Time) {
 		mlog.Infof("onHandshakeSuccess: %v", when)
-	}, nil)
+	})
 
 	// wait for initial reading
 	time.Sleep(2 * time.Second)
@@ -130,13 +66,14 @@ func TestWGWatcher_ReEnable(t *testing.T) {
 	watcher := NewWGWatcher(mlog, mocWgIface, "", newStateDump("peer", mlog, &Status{}))
 
 	ctx, cancel := context.WithCancel(context.Background())
-	watcher.PrepareInitialHandshake()
+	ok := watcher.PrepareInitialHandshake()
+	require.True(t, ok, "watcher should not be enabled yet")
 
 	wg := &sync.WaitGroup{}
 	wg.Add(1)
 	go func() {
 		defer wg.Done()
-		watcher.EnableWgWatcher(ctx, time.Now(), func() {}, func(when time.Time) {}, nil)
+		watcher.EnableWgWatcher(ctx, time.Now(), func() {}, func(when time.Time) {})
 	}()
 	cancel()
 
@@ -146,12 +83,13 @@ func TestWGWatcher_ReEnable(t *testing.T) {
 	ctx, cancel = context.WithCancel(context.Background())
 	defer cancel()
 
-	watcher.PrepareInitialHandshake()
+	ok = watcher.PrepareInitialHandshake()
+	require.True(t, ok, "watcher should be re-enabled after the previous run stopped")
 
 	onDisconnected := make(chan struct{}, 1)
 	go watcher.EnableWgWatcher(ctx, time.Now(), func() {
 		onDisconnected <- struct{}{}
-	}, func(when time.Time) {}, nil)
+	}, func(when time.Time) {})
 
 	time.Sleep(2 * time.Second)
 	mocWgIface.disconnect()

client/internal/profilemanager/config.go

@@ -96,7 +96,6 @@ type ConfigInput struct {
 	BlockLANAccess      *bool
 	BlockInbound        *bool
 	DisableIPv6         *bool
-	SyncMessageVersion  *int
 
 	DisableNotifications *bool
 
@@ -138,7 +137,6 @@ type Config struct {
 	BlockLANAccess      bool
 	BlockInbound        bool
 	DisableIPv6         bool
-	SyncMessageVersion  *int
 
 	DisableNotifications *bool
 
@@ -589,12 +587,6 @@ func (config *Config) apply(input ConfigInput) (updated bool, err error) {
 		updated = true
 	}
 
-	if input.SyncMessageVersion != nil && *input.SyncMessageVersion != *config.SyncMessageVersion {
-		log.Infof("setting SyncMessageVersion to %v", *input.SyncMessageVersion)
-		*config.SyncMessageVersion = *input.SyncMessageVersion
-		updated = true
-	}
-
 	if input.DisableNotifications != nil && (config.DisableNotifications == nil || *input.DisableNotifications != *config.DisableNotifications) {
 		if *input.DisableNotifications {
 			log.Infof("disabling notifications")

client/internal/profilemanager/service.go

@@ -11,7 +11,6 @@ import (
 	"runtime"
 	"sort"
 	"strings"
-	"syscall"
 
 	log "github.com/sirupsen/logrus"
 
@@ -440,11 +439,7 @@ func (s *ServiceManager) GetStatePath() string {
 
 	activeProf, err := s.GetActiveProfileState()
 	if err != nil {
-		if errors.Is(err, syscall.ENOSYS) {
+		log.Warnf("failed to get active profile state: %v", err)
-			log.Debugf("active profile state unavailable on this platform: %v", err)
-		} else {
-			log.Warnf("failed to get active profile state: %v", err)
-		}
 		return defaultStatePath
 	}
 

client/internal/profilemanager/state.go

@@ -4,7 +4,6 @@ import (
 	"context"
 	"errors"
 	"fmt"
-	"os"
 	"path/filepath"
 
 	"github.com/netbirdio/netbird/util"
@@ -72,22 +71,3 @@ func (pm *ProfileManager) SetActiveProfileState(state *ProfileState) error {
 
 	return nil
 }
-
-// RemoveProfileState deletes the per-profile state file (which holds the
-// account email used for the SSO login hint and the UI display). Called after
-// a successful logout so a logged-out profile no longer shows a stale account
-// email. The state file only stores the email, so deleting it is equivalent to
-// clearing it; the next SSO login recreates it. A missing file is not an error.
-func (pm *ProfileManager) RemoveProfileState(profileName string) error {
-	configDir, err := getConfigDir()
-	if err != nil {
-		return fmt.Errorf("get config directory: %w", err)
-	}
-
-	stateFile := filepath.Join(configDir, profileName+".state.json")
-	if err := os.Remove(stateFile); err != nil && !os.IsNotExist(err) {
-		return fmt.Errorf("remove profile state: %w", err)
-	}
-
-	return nil
-}

client/internal/rosenpass/manager.go

@@ -39,7 +39,6 @@ type rpServer interface {
 
 type Manager struct {
 	ifaceName    string
-	localWgKey   wgtypes.Key
 	spk          []byte
 	ssk          []byte
 	rpKeyHash    string
@@ -52,9 +51,8 @@ type Manager struct {
 	wgIface      PresharedKeySetter
 }
 
-// NewManager creates a new Rosenpass manager. localWgKey is the local
+// NewManager creates a new Rosenpass manager
-// WireGuard public key, used to derive the per-peer rendezvous key.
+func NewManager(preSharedKey *wgtypes.Key, wgIfaceName string) (*Manager, error) {
-func NewManager(preSharedKey *wgtypes.Key, wgIfaceName string, localWgKey wgtypes.Key) (*Manager, error) {
 	public, secret, err := rp.GenerateKeyPair()
 	if err != nil {
 		return nil, err
@@ -64,7 +62,6 @@ func NewManager(preSharedKey *wgtypes.Key, wgIfaceName string, localWgKey wgtype
 	log.Tracef("generated new rosenpass key pair with public key %s", rpKeyHash)
 	return &Manager{
 		ifaceName:    wgIfaceName,
-		localWgKey:   localWgKey,
 		rpKeyHash:    rpKeyHash,
 		spk:          public,
 		ssk:          secret,
@@ -76,7 +73,7 @@ func NewManager(preSharedKey *wgtypes.Key, wgIfaceName string, localWgKey wgtype
 		// nil receiver in addPeer -> m.rpWgHandler.AddPeer. generateConfig will
 		// replace it with a fresh handler on each Run() to clear stale peer
 		// state from previous engine sessions.
-		rpWgHandler: NewNetbirdHandler((*[32]byte)(preSharedKey), localWgKey),
+		rpWgHandler: NewNetbirdHandler(),
 		lock:        sync.Mutex{},
 	}, nil
 }
@@ -164,7 +161,7 @@ func (m *Manager) generateConfig() (rp.Config, error) {
 	cfg.Peers = []rp.PeerConfig{}
 
 	m.lock.Lock()
-	m.rpWgHandler = NewNetbirdHandler(m.preSharedKey, m.localWgKey)
+	m.rpWgHandler = NewNetbirdHandler()
 	if m.wgIface != nil {
 		m.rpWgHandler.SetInterface(m.wgIface)
 	}

client/internal/rosenpass/manager_test.go

@@ -85,7 +85,7 @@ func newTestManager(spkFirstByte byte, mock *mockServer) *Manager {
 		ssk:         make([]byte, 32),
 		rpKeyHash:   "test-hash",
 		rpPeerIDs:   make(map[string]*rp.PeerID),
-		rpWgHandler: NewNetbirdHandler(nil, wgtypes.Key{0x01}),
+		rpWgHandler: NewNetbirdHandler(),
 		server:      mock,
 	}
 }
@@ -255,7 +255,7 @@ func TestAddPeer_NilServer_ReturnsErrorNoCrash(t *testing.T) {
 // issue #4341 cannot occur in the window between NewManager and Run().
 func TestNewManager_PreInitializesHandler(t *testing.T) {
 	psk := wgtypes.Key{}
-	m, err := NewManager(&psk, "wt0", wgtypes.Key{0x01})
+	m, err := NewManager(&psk, "wt0")
 	require.NoError(t, err)
 	require.NotNil(t, m.rpWgHandler, "rpWgHandler must be initialized in NewManager")
 }
@@ -329,10 +329,10 @@ func TestIsPresharedKeyInitialized_AddedButNotHandshaken_ReturnsFalse(t *testing
 	require.False(t, m.IsPresharedKeyInitialized(wgKey))
 }
 
-// --- NetbirdHandler.applyKey ----------------------------------------------
+// --- NetbirdHandler.outputKey ----------------------------------------------
 
-func TestHandler_ApplyKey_FirstCallUsesUpdateOnlyFalse(t *testing.T) {
+func TestHandler_OutputKey_FirstCallUsesUpdateOnlyFalse(t *testing.T) {
-	h := NewNetbirdHandler(nil, wgtypes.Key{0x01})
+	h := NewNetbirdHandler()
 	iface := &mockIface{}
 	h.SetInterface(iface)
 
@@ -348,8 +348,8 @@ func TestHandler_ApplyKey_FirstCallUsesUpdateOnlyFalse(t *testing.T) {
 	require.Equal(t, wgKey.String(), iface.calls[0].peerKey)
 }
 
-func TestHandler_ApplyKey_SubsequentCallsUseUpdateOnlyTrue(t *testing.T) {
+func TestHandler_OutputKey_SubsequentCallsUseUpdateOnlyTrue(t *testing.T) {
-	h := NewNetbirdHandler(nil, wgtypes.Key{0x01})
+	h := NewNetbirdHandler()
 	iface := &mockIface{}
 	h.SetInterface(iface)
 
@@ -364,8 +364,8 @@ func TestHandler_ApplyKey_SubsequentCallsUseUpdateOnlyTrue(t *testing.T) {
 	require.True(t, iface.calls[1].updateOnly, "subsequent rotations must use updateOnly=true")
 }
 
-func TestHandler_ApplyKey_NilInterface_NoCrashNoCall(t *testing.T) {
+func TestHandler_OutputKey_NilInterface_NoCrashNoCall(t *testing.T) {
-	h := NewNetbirdHandler(nil, wgtypes.Key{0x01})
+	h := NewNetbirdHandler()
 	// no SetInterface — iface remains nil
 	pid := rp.PeerID{0x03}
 	h.AddPeer(pid, "wt0", rp.Key(wgtypes.Key{}))
@@ -374,8 +374,8 @@ func TestHandler_ApplyKey_NilInterface_NoCrashNoCall(t *testing.T) {
 	h.HandshakeCompleted(pid, rp.Key{})
 }
 
-func TestHandler_ApplyKey_UnknownPeer_NoCall(t *testing.T) {
+func TestHandler_OutputKey_UnknownPeer_NoCall(t *testing.T) {
-	h := NewNetbirdHandler(nil, wgtypes.Key{0x01})
+	h := NewNetbirdHandler()
 	iface := &mockIface{}
 	h.SetInterface(iface)
 
@@ -384,7 +384,7 @@ func TestHandler_ApplyKey_UnknownPeer_NoCall(t *testing.T) {
 }
 
 func TestHandler_RemovePeer_ClearsInitializedState(t *testing.T) {
-	h := NewNetbirdHandler(nil, wgtypes.Key{0x01})
+	h := NewNetbirdHandler()
 	iface := &mockIface{}
 	h.SetInterface(iface)
 
@@ -398,7 +398,7 @@ func TestHandler_RemovePeer_ClearsInitializedState(t *testing.T) {
 }
 
 func TestHandler_SetInterfaceAfterAddPeer_StillReceivesKey(t *testing.T) {
-	h := NewNetbirdHandler(nil, wgtypes.Key{0x01})
+	h := NewNetbirdHandler()
 	pid := rp.PeerID{0x05}
 	wgKey := wgtypes.Key{0xEE}
 	h.AddPeer(pid, "wt0", rp.Key(wgKey))