Merge pull request #2 from MedUnes/feat/support-desktop

feat(support desktop)

Author: MedUnes

.github/workflows/test.yml

@@ -46,7 +46,9 @@ jobs:
     strategy:
       fail-fast: false
       matrix:
-        os: [noble, jammy, focal, trixie, bookworm, bullseye]
+        os: [noble, jammy, focal, trixie, bookworm, bullseye,
+             noble-desktop, jammy-desktop, focal-desktop,
+             trixie-desktop, bookworm-desktop, bullseye-desktop]
     steps:
       - uses: actions/checkout@v4
 

Makefile

@@ -12,4 +12,18 @@ test-systemd: build
 	./test/systemd/build-local-deb.sh bin/docker-dns 0.0.0-local
 	./test/systemd/test-systemd.sh $(OS)
 
-.PHONY: build test run test-systemd
+test-systemd-all: build
+	@for os in noble jammy focal trixie bookworm bullseye noble-desktop jammy-desktop focal-desktop trixie-desktop bookworm-desktop bullseye-desktop; do \
+		echo "=== Testing $$os ==="; \
+		./test/systemd/build-local-deb.sh bin/docker-dns 0.0.0-local && \
+		./test/systemd/test-systemd.sh $$os || echo "FAILED: $$os"; \
+	done
+
+test-systemd-desktop-all: build
+	@for os in noble-desktop jammy-desktop focal-desktop trixie-desktop bookworm-desktop bullseye-desktop; do \
+		echo "=== Testing $$os ==="; \
+		./test/systemd/build-local-deb.sh bin/docker-dns 0.0.0-local && \
+		./test/systemd/test-systemd.sh $$os || echo "FAILED: $$os"; \
+	done
+
+.PHONY: build test run test-systemd test-systemd-all test-systemd-desktop-all

README.md

@@ -11,12 +11,15 @@ between containers and the host.
 
 ## Features
 
-- **Automatic DNS Resolution**: Resolve Docker container names with a custom TLD to their respective IP addresses.
-- **Caching Mechanism**: Configurable TTL for DNS cache entries to improve performance.
-- **Simple Configuration**: Minimal setup with command-line options or a configuration file.
-- **Debian Package**: Easy-to-install `.deb` package for seamless integration.
-- **Fallback DNS**: Forwards non-Docker queries to configurable DNS servers.
-- **Lightweight and Fast**: Built with Go for high performance.
+- **Automatic DNS Resolution**: Resolve Docker container names with a custom TLD (default `.docker`) to their IP addresses. Supports multiple TLDs and containers on any Docker network.
+- **Fallback DNS**: Forwards non-Docker queries in parallel to configurable upstream resolvers (default: `8.8.8.8`, `1.1.1.1`, `8.8.4.4`), returning the first successful response.
+- **Caching**: TTL-based DNS cache with background eviction, size limits, and hit/miss telemetry.
+- **Rate Limiting**: Per-IP token-bucket rate limiter with automatic idle cleanup.
+- **Health & Metrics**: HTTP server on `:8080` exposes `/health` and `/metrics` (cache stats, query counts, error rates).
+- **UDP + TCP**: Full DNS protocol support with EDNS0 handling and proper truncation.
+- **Debian Package**: `.deb` package with automatic systemd integration and clean uninstall.
+- **Tested on 12 Configurations**: Full install → resolve → uninstall lifecycle CI on Ubuntu 20.04/22.04/24.04 and Debian 11/12/13, both server and desktop variants.
+- **Lightweight**: Single Go binary, minimal resource footprint.
 
 ## Architecture
 
@@ -146,23 +149,18 @@ integrates with DNS resolution for each ditro/version..
 
 ---
 
-## Important Notes
+## DNS Integration
 
-### Default Resolver Integration
+The `.deb` package auto-detects your system's DNS resolver and integrates accordingly:
 
-- While Docker DNS runs by default on a custom IP (not `127.0.0.1`), it is possible to use tools like `dig` to test
-  queries:
-  ```bash
-       dig mycontainer.docker @127.0.0.153 +short 
-  ```
-    - However, making Docker DNS the system-wide default resolver requires additional configuration or hacks.
+- **Ubuntu** (systemd-resolved): routing domain drop-in at `/etc/systemd/resolved.conf.d/docker-dns.conf` — only
+  `.docker` queries go to docker-dns, everything else is untouched
+- **Debian Desktop** (NetworkManager): dispatcher script at `/etc/NetworkManager/dispatcher.d/docker-dns` re-prepends
+  the nameserver after every NM event
+- **Debian Server** (plain resolv.conf): prepends `nameserver 127.0.0.153` to `/etc/resolv.conf`
 
-### Systemd-Resolved Compatibility: Direct integration with
-`systemd-resolved` is non-trivial and not recommended without advanced setup.
-
-### Protocol
-
-* `docker-dns` is configured to run on `UDP` mode only, so `TCP` requests won't be answered
+See [docs/Systemd.md](./docs/Systemd.md) for the full details, browser DNS-over-HTTPS caveats, and manual integration
+examples for custom resolvers (dnsmasq, unbound, Pi-hole).
 
 ---
 
@@ -192,17 +190,31 @@ integrates with DNS resolution for each ditro/version..
 - Once you have the binary generated, you can run at as follows (linux):
    ```bash
    sudo ./docker-dns -h
-      Usage of docker-dns:
-   -ip string
-      IP address the DNS server listens on (default "127.0.0.153")
-   -tld string
-      Comma-separated managed top-level domains for container resolution (default "docker")
-   -ttl int
-      TTL in seconds for cache entries and DNS responses (default 300)
-   -resolvers string
-      Comma-separated fallback DNS resolver IPs (default "8.8.8.8,1.1.1.1,8.8.4.4")
-   -forward-timeout duration
-      Per-resolver timeout for forwarded DNS queries (default 2s)
+   Usage of docker-dns:
+     -ip string
+         IP address the DNS server listens on (default "127.0.0.153")
+     -tld string
+         Comma-separated managed top-level domains for container resolution (default "docker")
+     -ttl int
+         TTL in seconds for cache entries and DNS responses (default 300)
+     -resolvers string
+         Comma-separated fallback DNS resolver IPs (default "8.8.8.8,1.1.1.1,8.8.4.4")
+     -forward-timeout duration
+         Per-resolver timeout for forwarded DNS queries (default 2s)
+     -rate-limit float
+         Max queries/sec per client IP; 0 disables rate limiting (default 100)
+     -rate-burst int
+         Burst allowance for per-IP rate limiting (default 50)
+     -max-cache-size int
+         Max DNS cache entries; 0 = unlimited (default 10000)
+     -http-addr string
+         Address for the health/metrics HTTP server; empty to disable (default ":8080")
+     -docker-timeout duration
+         Timeout for Docker API calls (default 5s)
+     -docker-host string
+         Docker host override (empty = use DOCKER_HOST env / socket default)
+     -log-level string
+         Log level: debug | info | warn | error (default "info")
    ```
 - P.S: `sudo` (or `root`) is required as the server will be listening on port `53`, which is
   a [previewed port](https://www.w3.org/Daemon/User/Installation/PrivilegedPorts.html)

build/dpkg/DEBIAN/postinst

@@ -57,6 +57,16 @@ configure)
     done
     ROUTING_DOMAINS=$(echo "$ROUTING_DOMAINS" | xargs) # trim leading space
 
+    # Prepend `nameserver $CUSTOM_IP` to /etc/resolv.conf (idempotent).
+    prepend_resolv_conf() {
+        grep -q "$CUSTOM_IP" /etc/resolv.conf 2>/dev/null && \
+            echo "detected an entry of $CUSTOM_IP in /etc/resolv.conf, removing.." && \
+            grep -v "$CUSTOM_IP" /etc/resolv.conf > /tmp/resolv.tmp && \
+            cat /tmp/resolv.tmp > /etc/resolv.conf && rm -f /tmp/resolv.tmp
+        { printf '# Generated by docker-dns\nnameserver %s\n' "$CUSTOM_IP"; cat /etc/resolv.conf; } > /tmp/resolv.tmp \
+            && cat /tmp/resolv.tmp > /etc/resolv.conf && rm -f /tmp/resolv.tmp
+    }
+
     if systemctl is-active systemd-resolved.service >/dev/null 2>&1; then
         echo "Detected systemd-resolved running. Extending its configuration.."
 
@@ -72,15 +82,41 @@ Domains=$ROUTING_DOMAINS
 EOF
         systemctl restart systemd-resolved.service
         echo "systemd-resolved extended with DNS=$CUSTOM_IP and Domains=$ROUTING_DOMAINS."
+    elif systemctl is-active NetworkManager.service >/dev/null 2>&1; then
+        # NetworkManager active without systemd-resolved (Debian Desktop).
+        # NM rewrites /etc/resolv.conf on every connection event and would clobber
+        # a one-shot prepend. Install a dispatcher script so NM re-prepends for us
+        # after every up/down/dhcp event, and prepend once immediately so DNS
+        # works between now and the next NM event.
+        echo "Detected NetworkManager without systemd-resolved. Installing dispatcher.."
+
+        mkdir -p /etc/NetworkManager/dispatcher.d
+        cat <<EOF >/etc/NetworkManager/dispatcher.d/docker-dns
+#!/bin/bash
+# Re-prepend docker-dns nameserver after NetworkManager rewrites /etc/resolv.conf.
+# NM invokes dispatchers with "\$1 = interface" and "\$2 = action".
+# up/dhcp*-change: NM has just written resolv.conf with connection-provided DNS.
+# down: NM rewrites resolv.conf to drop the departing interface's DNS servers.
+case "\$2" in
+    up|down|dhcp4-change|dhcp6-change)
+        if ! grep -q '$CUSTOM_IP' /etc/resolv.conf 2>/dev/null; then
+            { printf '# Generated by docker-dns\nnameserver %s\n' '$CUSTOM_IP'
+              cat /etc/resolv.conf
+            } > /tmp/resolv.docker-dns.tmp \\
+            && cat /tmp/resolv.docker-dns.tmp > /etc/resolv.conf \\
+            && rm -f /tmp/resolv.docker-dns.tmp
+        fi
+        ;;
+esac
+EOF
+        chmod 0755 /etc/NetworkManager/dispatcher.d/docker-dns
+        chown root:root /etc/NetworkManager/dispatcher.d/docker-dns
+
+        prepend_resolv_conf
+        echo "NetworkManager dispatcher installed; /etc/resolv.conf prepended with nameserver $CUSTOM_IP."
     else
-        echo "systemd-resolved is not active. Updating /etc/resolv.conf.."
-        grep -q "$CUSTOM_IP" /etc/resolv.conf 2>/dev/null && \
-            echo "detected an entry of $CUSTOM_IP in /etc/resolv.conf, removing.." && \
-            grep -v "$CUSTOM_IP" /etc/resolv.conf > /tmp/resolv.tmp && \
-            cat /tmp/resolv.tmp > /etc/resolv.conf && rm -f /tmp/resolv.tmp
-        # Prepend so docker-dns is queried before other nameservers
-        { printf '# Generated by docker-dns\nnameserver %s\n' "$CUSTOM_IP"; cat /etc/resolv.conf; } > /tmp/resolv.tmp \
-            && cat /tmp/resolv.tmp > /etc/resolv.conf && rm -f /tmp/resolv.tmp
+        echo "Neither systemd-resolved nor NetworkManager is active. Updating /etc/resolv.conf.."
+        prepend_resolv_conf
         echo "/etc/resolv.conf updated with nameserver $CUSTOM_IP."
     fi
     ;;

build/dpkg/DEBIAN/postrm

@@ -22,6 +22,10 @@ remove|purge)
     rm -f /etc/unbound/unbound.conf.d/docker-dns.conf
     systemctl restart unbound 2>/dev/null || true
 
+    # NetworkManager dispatcher — remove BEFORE restarting NM so a restart event
+    # cannot fire the dispatcher and re-add the nameserver to resolv.conf.
+    rm -f /etc/NetworkManager/dispatcher.d/docker-dns
+
     # NetworkManager dnsmasq
     rm -f /etc/NetworkManager/dnsmasq.d/docker-dns.conf
     systemctl restart NetworkManager 2>/dev/null || true

docs/Systemd.md

@@ -86,31 +86,69 @@ The `~` prefix on `docker` makes it a routing domain, not a search domain. This
 
 This is the same mechanism used by VPNs, LXD, and corporate split-DNS setups.
 
-### On systems without systemd-resolved (Debian default)
+### On systems with NetworkManager (Debian Desktop)
 
-The postinst appends docker-dns as a nameserver to `/etc/resolv.conf`:
+On Debian Desktop, NetworkManager is active but systemd-resolved is not installed. NM runs with `dns=default`, meaning
+it writes `/etc/resolv.conf` directly on every connection event, DHCP renewal, or service restart. A one-shot prepend
+to `/etc/resolv.conf` would be overwritten the next time NM touches the file.
+
+The postinst detects this (NM active, resolved not active) and installs a dispatcher script:
+
+```bash
+# /etc/NetworkManager/dispatcher.d/docker-dns
+```
+
+The dispatcher re-prepends `nameserver 127.0.0.153` to `/etc/resolv.conf` after every NM connection event (`up`,
+`down`, `dhcp4-change`, `dhcp6-change`). The postinst also does an immediate one-shot prepend so DNS works between
+install and the next NM event.
+
+The postrm removes the dispatcher script and cleans up `/etc/resolv.conf` on uninstall.
+
+### On systems without systemd-resolved or NetworkManager (Debian Server)
+
+On minimal or server installs where neither systemd-resolved nor NetworkManager is active, the postinst prepends
+docker-dns as a nameserver to `/etc/resolv.conf`:
 
 ```
 # Generated by docker-dns
 nameserver 127.0.0.153
 ```
 
-Since docker-dns handles `.docker` queries itself and forwards everything else to its configured upstream resolvers (
-default: 8.8.8.8, 1.1.1.1, 8.8.4.4), this works for both container resolution and normal DNS.
+Since docker-dns handles `.docker` queries itself and forwards everything else to its configured upstream resolvers
+(default: 8.8.8.8, 1.1.1.1, 8.8.4.4), this works for both container resolution and normal DNS.
 
 The postrm removes this entry on uninstall using the `# Generated by docker-dns` comment as a marker.
 
-### On systems with other resolvers
+### Custom DNS Resolvers (Manual Configuration)
 
-The postinst detects the active resolver and writes the appropriate configuration:
+docker-dns's automatic integration covers the default DNS stack on all supported Ubuntu and Debian versions (server and
+desktop). If you run a custom local resolver (dnsmasq, unbound, Pi-hole, or NM with `dns=dnsmasq`), you already know
+your DNS stack — add the forwarding rule manually.
 
-| Resolver                      | Config file                                     | Directive                           |
-|-------------------------------|-------------------------------------------------|-------------------------------------|
-| dnsmasq                       | `/etc/dnsmasq.d/docker-dns.conf`                | `server=/docker/127.0.0.153`        |
-| unbound                       | `/etc/unbound/unbound.conf.d/docker-dns.conf`   | `stub-zone` pointing to 127.0.0.153 |
-| NetworkManager (dnsmasq mode) | `/etc/NetworkManager/dnsmasq.d/docker-dns.conf` | `server=/docker/127.0.0.153`        |
+**dnsmasq** (standalone or Pi-hole):
+```
+# /etc/dnsmasq.d/docker-dns.conf
+server=/docker/127.0.0.153
+```
+Then: `sudo systemctl restart dnsmasq`
+
+**unbound**:
+```yaml
+# /etc/unbound/unbound.conf.d/docker-dns.conf
+stub-zone:
+    name: "docker"
+    stub-addr: 127.0.0.153
+```
+Then: `sudo systemctl restart unbound`
 
-All of these are cleaned up by the postrm on uninstall.
+**NetworkManager with dns=dnsmasq**:
+```
+# /etc/NetworkManager/dnsmasq.d/docker-dns.conf
+server=/docker/127.0.0.153
+```
+Then: `sudo systemctl restart NetworkManager`
+
+Note: the postrm does not clean up manually-created configs. If you uninstall docker-dns, remove these files yourself.
 
 ---
 
@@ -187,12 +225,18 @@ systemctl restart dnsmasq 2>/dev/null || true
 rm -f /etc/unbound/unbound.conf.d/docker-dns.conf
 systemctl restart unbound 2>/dev/null || true
 
+# NetworkManager dispatcher — removed before NM restart so a restart event
+# cannot fire the dispatcher and re-add the nameserver to resolv.conf
+rm -f /etc/NetworkManager/dispatcher.d/docker-dns
+
 # NetworkManager dnsmasq
 rm -f /etc/NetworkManager/dnsmasq.d/docker-dns.conf
 systemctl restart NetworkManager 2>/dev/null || true
 
-# Plain resolv.conf: remove the appended entry
-sed -i '/# Generated by docker-dns/,+1d' /etc/resolv.conf
+# Plain resolv.conf: remove the prepended entry
+awk '/# Generated by docker-dns/{skip=1; next} skip{skip=0; next} 1' \
+    /etc/resolv.conf > /tmp/resolv.tmp \
+    && cat /tmp/resolv.tmp > /etc/resolv.conf && rm -f /tmp/resolv.tmp
 ```
 
 Each `systemctl restart` is guarded with `2>/dev/null || true` so it silently skips services that aren't installed.
@@ -205,14 +249,22 @@ The systemd integration tests (`make test-systemd OS=<codename>`) verify the ful
 lifecycle on each target. Both the direct path (`dig @127.0.0.153`) and the system-level path (`dig` without `@`, going
 through whatever the OS uses for DNS) are tested.
 
-| OS       | Version      | DNS method                        | Assertions |
-|----------|--------------|-----------------------------------|------------|
-| Noble    | Ubuntu 24.04 | systemd-resolved + routing domain | 13         |
-| Jammy    | Ubuntu 22.04 | systemd-resolved + routing domain | 13         |
-| Focal    | Ubuntu 20.04 | systemd-resolved + routing domain | 13         |
-| Trixie   | Debian 13    | plain resolv.conf                 | 12         |
-| Bookworm | Debian 12    | plain resolv.conf                 | 12         |
-| Bullseye | Debian 11    | plain resolv.conf                 | 12         |
+| OS       | Version      | Variant | DNS Method                                      | Assertions |
+|----------|--------------|---------|-------------------------------------------------|------------|
+| Noble    | Ubuntu 24.04 | Server  | systemd-resolved + routing domain               | 13         |
+| Noble    | Ubuntu 24.04 | Desktop | systemd-resolved + NM (dns=systemd-resolved)    | 14         |
+| Jammy    | Ubuntu 22.04 | Server  | systemd-resolved + routing domain               | 13         |
+| Jammy    | Ubuntu 22.04 | Desktop | systemd-resolved + NM (dns=systemd-resolved)    | 14         |
+| Focal    | Ubuntu 20.04 | Server  | systemd-resolved + routing domain               | 13         |
+| Focal    | Ubuntu 20.04 | Desktop | systemd-resolved + NM (dns=systemd-resolved)    | 14         |
+| Trixie   | Debian 13    | Server  | plain resolv.conf                               | 12         |
+| Trixie   | Debian 13    | Desktop | NM dispatcher + resolv.conf                     | 15         |
+| Bookworm | Debian 12    | Server  | plain resolv.conf                               | 12         |
+| Bookworm | Debian 12    | Desktop | NM dispatcher + resolv.conf                     | 15         |
+| Bullseye | Debian 11    | Server  | plain resolv.conf                               | 12         |
+| Bullseye | Debian 11    | Desktop | NM dispatcher + resolv.conf                     | 15         |
+
+Desktop variants additionally test that docker-dns resolution survives a full `systemctl restart NetworkManager` cycle.
 
 Each test runs inside a privileged Docker container with systemd as PID 1, Docker-in-Docker for container resolution,
 and full DNS chain verification including post-uninstall cleanup.