A rootless Android app that boots Alpine Linux: run containers (Podman/Docker/LXC) and GUI desktop apps.
docs: note udp port forwarding works on both backends
ExTvfd4eac84cf0b
diff --git a/docs/guide/limitations.html b/docs/guide/limitations.htmlindex cfd30ec..43dc91d 100644--- a/docs/guide/limitations.html+++ b/docs/guide/limitations.html@@ -81,11 +81,6 @@<td>Pixel pKVM devices only; two ADB permissions required</td><td>AVF uses the Android Virtualization Framework, which requires a device that ships pKVM (currently Pixel 8 and later). Two one-time <code>adb</code> grants (or Shizuku) are needed before AVF can start. Devices that only support protected VMs and not non-protected VMs cannot use the AVF backend at all; Podroid falls back to QEMU automatically.</td></tr>- <tr>- <td>AVF networking</td>- <td>TCP port forwarding only</td>- <td>The AVF backend forwards TCP ports via the in-app port-forward rules. UDP rules are silently ignored when AVF is active. If you need UDP (e.g. DNS on port 53 serving LAN clients), switch to QEMU.</td>- </tr><tr><td>AVF CPU count</td><td>No fine-grained vCPU selection</td>diff --git a/docs/guide/networking.html b/docs/guide/networking.htmlindex 7b1bfd4..0d5a6b7 100644--- a/docs/guide/networking.html+++ b/docs/guide/networking.html@@ -134,7 +134,7 @@ podroid-forward 8888 80podroid-forward list<span class="c"># drop a rule</span>podroid-forward remove 8888 tcp</code></pre>- <p>Rules created this way are identical to ones added in Settings: they apply live, persist across reboots, and appear in the Settings list. UDP is QEMU-only (ignored on AVF), the same as the UI.</p>+ <p>Rules created this way are identical to ones added in Settings: they apply live, persist across reboots, and appear in the Settings list. Both TCP and UDP work on either backend.</p><h3><code>podroid-notify</code> - post an Android notification</h3><pre><code><span class="c"># simplest form</span>@@ -176,7 +176,7 @@ podroid-notify --id 7 "progress 50%"</code></pre><div class="callout warn"><span class="label">Workaround</span>- <p>Forward to a high host port instead of the standard one: map host <code>8080</code> to guest <code>80</code>, and host <code>8443</code> to guest <code>443</code>. Services inside the guest can still listen on port 80 or 443 as usual - only the host-side port is restricted.</p>+ <p>Forward to a high host port instead of the standard one: map host <code>8080</code> to guest <code>80</code>, and host <code>8443</code> to guest <code>443</code>. A daemon running as guest-root can still listen on 80 or 443. A <strong>rootless Podman or Docker</strong> container is the exception: its published port is also subject to the ≥1024 limit inside the guest, so publish to a high guest port too - e.g. a Pi-hole DNS container on guest <code>5353/udp</code>, forwarded to host <code>5353</code>, with LAN clients pointed at <code><phone-ip>:5353</code>.</p></div><pre><code><span class="c"># Forward a web server on guest :80 to a high host port</span>@@ -186,11 +186,11 @@ podroid-notify --id 7 "progress 50%"</code></pre><span class="p">$</span> curl http://<phone-ip>:8080</code></pre><h2>Port forwarding on AVF</h2>- <p>On the AVF backend, port forwarding is implemented over <strong>vsock</strong> rather than SLIRP hostfwd. Each rule creates a TCP listener on <code>0.0.0.0:hostPort</code> on the Android side, which is bridged to a vsock port on the guest via the <code>podroid-vsock-agent</code> service. The privileged-port limit applies here too, for the same reason.</p>+ <p>On the AVF backend, port forwarding is implemented over <strong>vsock</strong> rather than SLIRP hostfwd. Each rule creates a listener on <code>0.0.0.0:hostPort</code> on the Android side, which is bridged to a vsock port on the guest via the <code>podroid-vsock-agent</code> service. The privileged-port limit applies here too, for the same reason.</p><div class="callout note"><span class="label">UDP on AVF</span>- <p>UDP rules are ignored on the AVF backend. The vsock transport is TCP-only. If you need UDP forwarding, use the QEMU backend.</p>+ <p>UDP forwarding works on both backends. Because vsock is a stream transport, each inbound UDP datagram is length-framed and carried over a per-client vsock connection, then delivered to the guest as a real UDP datagram - so a guest DNS server such as Pi-hole is reachable over UDP. Idle flows are reaped automatically, and the number of concurrent client flows is capped to bound resource use.</p></div><div class="pagenav">