18 Commits

Author SHA1 Message Date
Datong Sun
827530f62c chore(phantun) release v0.4.0 and bump fake-tcp dependency to v0.4 2022-04-16 03:48:07 -07:00
Datong Sun
245cb9c7f4 chore(fake-tcp) release v0.4.0 2022-04-16 03:47:02 -07:00
Datong Sun
85555f2a34 feat(*) IPv6 support 2022-04-16 18:43:55 +08:00
Datong Sun
74183071f1 style(phantun) remove unnecessary tokio::select call 2022-04-15 23:01:44 +08:00
Datong Sun
2f4eaafccd docs(README) fixed a typo 2022-04-10 23:40:44 +08:00
Datong Sun
1e3b632413 docs(README) add benchmarking results based on v0.3.2 2022-04-10 21:21:09 +08:00
Datong Sun
99bff568f6 chore(phantun) release v0.3.2 2022-04-10 06:10:57 -07:00
Datong Sun
91ad2c03a1 chore(fake-tcp) release v0.3.1 2022-04-10 06:09:48 -07:00
Datong Sun
581d80d08c perf(fake-tcp) use flume to avoid locking in receiver, improved single
connection performance by 300%
2022-04-10 21:07:12 +08:00
Datong Sun
55da4d6a62 docs(README) style improvements 2022-04-10 18:47:22 +08:00
Datong Sun
bb859be6b6 docs(README) add build status and docs.rs badge 2022-04-10 18:37:06 +08:00
Datong Sun
8d315ea4e7 docs(README) add packet header diagram 2022-04-10 18:26:28 +08:00
Datong Sun
21eabe8b82 docs(README) add description for safe Rust and bump latest release to
`v0.3.1`
2022-04-10 01:44:33 -07:00
Datong Sun
8a74b31c6e chore(phantun) bump fake-tcp dependency to v0.3.0 and release
`v0.3.1`
2022-04-10 01:37:59 -07:00
Datong Sun
ca14ba457f chore(fake-tcp) bump to v0.3.0 2022-04-10 01:36:45 -07:00
Datong Sun
33a0cfe567 docs(README) updated benchmarking results 2022-04-10 01:35:07 -07:00
Datong Sun
95dfd8ab54 fix(fake-tcp) fix an issue where RST generated is not following
the proper RFC requirement.

Send ACK every 128MB in lieu of data packets.
2022-04-10 16:33:53 +08:00
Datong Sun
1c35635091 docs(README) bump latest release version to v0.3.0 2022-04-09 08:49:29 -07:00
10 changed files with 545 additions and 287 deletions

View File

@@ -2,6 +2,9 @@
A lightweight and fast UDP to TCP obfuscator.
![GitHub Workflow Status](https://img.shields.io/github/workflow/status/dndx/phantun/Rust?style=flat-square)
![docs.rs](https://img.shields.io/docsrs/fake-tcp)
Table of Contents
=================
@@ -32,7 +35,7 @@ Table of Contents
# Latest release
[v0.2.5](https://github.com/dndx/phantun/releases/tag/v0.2.5)
[v0.3.2](https://github.com/dndx/phantun/releases/tag/v0.3.2)
# Overview
@@ -52,6 +55,11 @@ connection from the perspective of firewalls/NAT devices.
Phantun means Phantom TUN, as it is an obfuscator for UDP traffic that does just enough work
to make it pass through stateful firewall/NATs as TCP packets.
Phantun is written in 100% safe Rust. It has been optimized extensively to scale well on multi-core
systems and has no issue saturating all available CPU resources on a fast connection.
See the [Performance](#performance) section for benchmarking results.
![Phantun benchmark results](images/phantun-vs-udp2raw-benchmark-result.png)
![Traffic flow diagram](images/traffic-flow.png)
# Usage
@@ -218,18 +226,20 @@ RUST_LOG=info /usr/local/bin/phantun_client --local 127.0.0.1:1234 --remote exam
Phantun aims to keep tunneling overhead to the minimum. The overhead compared to a plain UDP packet
is the following:
Standard UDP packet: 20 byte IP header + 8 byte UDP header = 28 bytes
**Standard UDP packet:** `20 byte IP header + 8 byte UDP header = 28 bytes`
Phantun obfuscated UDP packet: 20 byte IP header + 20 byte TCP header = 40 bytes
**Obfuscated packet:** `20 byte IP header + 20 byte TCP header = 40 bytes`
Note that Phantun does not add any additional header other than IP and TCP headers in order to pass through
stateful packet inspection!
Phantun's additional overhead: 12 bytes. I other words, when using Phantun, the usable payload for
Phantun's additional overhead: `12 bytes`. I other words, when using Phantun, the usable payload for
UDP packet is reduced by 12 bytes. This is the minimum overhead possible when doing such kind
of obfuscation.
![Packet header diagram](images/packet-headers.png)
[Back to TOC](#table-of-contents)
## MTU calculation for WireGuard
@@ -237,14 +247,20 @@ of obfuscation.
For people who use Phantun to tunnel [WireGuard®](https://www.wireguard.com) UDP packets, here are some guidelines on figuring
out the correct MTU to use for your WireGuard interface.
```
WireGuard MTU = Interface MTU - IP header (20 bytes) - TCP header (20 bytes) - WireGuard overhead (32 bytes)
```
For example, for a Ethernet interface with 1500 bytes MTU, the WireGuard interface MTU should be set as:
```
1500 - 20 - 20 - 32 = 1428 bytes
```
The resulted Phantun TCP data packet will be 1500 bytes which does not exceed the
interface MTU of 1500.
interface MTU of 1500. Please note it is strongly recommended to use the same interface
MTU for both ends of a WireGuard tunnel, or unexected packet loss may occur and these issues are
generally very hard to troubleshoot.
[Back to TOC](#table-of-contents)
@@ -264,13 +280,21 @@ For users who wish to use `fake-tcp` library inside their own project, refer to
# Performance
Performance was tested on AWS t3.xlarge instance with 4 vCPUs and 5 Gb/s NIC. WireGuard was used
for tunneling TCP/UDP traffic between two test instances and MTU has been tuned to avoid fragmentation.
Performance was tested on 2 AWS `t4g.xlarge` instances with 4 vCPUs and 5 Gb/s NIC over LAN. `nftables` was used to redirect
UDP stream of `iperf3` to go through the Phantun/udp2raw tunnel between two test instances and MTU has been tuned to avoid fragmentation.
| | WireGuard | WireGuard + Phantun | WireGuard + udp2raw (cipher-mode=none auth-mode=none disable-anti-replay) |
|-----------------|-------------|---------------------|---------------------------------------------------------------------------|
| iperf3 -c IP -R | 1.56 Gbit/s | 540 Mbit/s | 369 Mbit/s |
| iperf3 -c IP | 1.71 Gbit/s | 519 Mbit/s | 312 Mbit/s |
Phantun `v0.3.2` and `udp2raw_arm_asm_aes` `20200818.0` was used. These were the latest release of both projects as of Apr 2022.
Test command: `iperf3 -c <IP> -p <PORT> -R -u -l 1400 -b 1000m -t 30 -P 5`
| Mode | Send Speed | Receive Speed | Overall CPU Usage |
|---------------------------------------------------------------------------------|----------------|----------------|-----------------------------------------------------|
| Direct (1 stream) | 3.00 Gbits/sec | 2.37 Gbits/sec | 25% (1 core at 100%) |
| Phantun (1 stream) | 1.30 Gbits/sec | 1.20 Gbits/sec | 60% (1 core at 100%, 3 cores at 50%) |
| udp2raw (`cipher-mode=none` `auth-mode=none` `disable-anti-replay`) (1 stream) | 1.30 Gbits/sec | 715 Mbits/sec | 40% (1 core at 100%, 1 core at 50%, 2 cores idling) |
| Direct connection (5 streams) | 5.00 Gbits/sec | 3.64 Gbits/sec | 25% (1 core at 100%) |
| Phantun (5 streams) | 5.00 Gbits/sec | 2.38 Gbits/sec | 95% (all cores utilized) |
| udp2raw (`cipher-mode=none` `auth-mode=none` `disable-anti-replay`) (5 streams) | 5.00 Gbits/sec | 770 Mbits/sec | 50% (2 cores at 100%) |
[Back to TOC](#table-of-contents)
@@ -300,7 +324,7 @@ Here is a quick overview of comparison between those two to help you choose:
| UDP over UDP obfuscation | ❌ | ✅ |
| Multi-threaded | ✅ | ❌ |
| Throughput | Better | Good |
| Raw IP mode | TUN interface | Raw sockets + BPF |
| Layer 3 mode | TUN interface | Raw sockets + BPF |
| Tunneling MTU overhead | 12 bytes | 44 bytes |
| Seprate TCP connections for each UDP connection | Client/Server | Server only |
| Anti-replay, encryption | ❌ | ✅ |

View File

@@ -1,6 +1,6 @@
[package]
name = "fake-tcp"
version = "0.2.4"
version = "0.4.0"
edition = "2021"
authors = ["Datong Sun <dndx@idndx.com>"]
license = "MIT OR Apache-2.0"
@@ -22,3 +22,4 @@ rand = { version = "0.8", features = ["small_rng"] }
log = "0.4"
internet-checksum = "0.2"
tokio-tun = "0.5"
flume = "0.10"

View File

@@ -49,27 +49,28 @@ use pnet::packet::{tcp, Packet};
use rand::prelude::*;
use std::collections::{HashMap, HashSet};
use std::fmt;
use std::net::{Ipv4Addr, SocketAddrV4};
use std::net::{IpAddr, Ipv4Addr, Ipv6Addr, SocketAddr};
use std::sync::atomic::{AtomicU32, Ordering};
use std::sync::{Arc, RwLock};
use tokio::sync::broadcast;
use tokio::sync::mpsc::{self, Receiver, Sender};
use tokio::sync::Mutex as AsyncMutex;
use tokio::sync::mpsc;
use tokio::time;
use tokio_tun::Tun;
const TIMEOUT: time::Duration = time::Duration::from_secs(1);
const RETRIES: usize = 6;
const MPSC_BUFFER_LEN: usize = 512;
const MPMC_BUFFER_LEN: usize = 512;
const MPSC_BUFFER_LEN: usize = 128;
const MAX_UNACKED_LEN: u32 = 128 * 1024 * 1024; // 128MB
#[derive(Hash, Eq, PartialEq, Clone, Debug)]
struct AddrTuple {
local_addr: SocketAddrV4,
remote_addr: SocketAddrV4,
local_addr: SocketAddr,
remote_addr: SocketAddr,
}
impl AddrTuple {
fn new(local_addr: SocketAddrV4, remote_addr: SocketAddrV4) -> AddrTuple {
fn new(local_addr: SocketAddr, remote_addr: SocketAddr) -> AddrTuple {
AddrTuple {
local_addr,
remote_addr,
@@ -78,17 +79,18 @@ impl AddrTuple {
}
struct Shared {
tuples: RwLock<HashMap<AddrTuple, Sender<Bytes>>>,
tuples: RwLock<HashMap<AddrTuple, flume::Sender<Bytes>>>,
listening: RwLock<HashSet<u16>>,
tun: Vec<Arc<Tun>>,
ready: Sender<Socket>,
ready: mpsc::Sender<Socket>,
tuples_purge: broadcast::Sender<AddrTuple>,
}
pub struct Stack {
shared: Arc<Shared>,
local_ip: Ipv4Addr,
ready: Receiver<Socket>,
local_ip6: Option<Ipv6Addr>,
ready: mpsc::Receiver<Socket>,
}
pub enum State {
@@ -101,11 +103,12 @@ pub enum State {
pub struct Socket {
shared: Arc<Shared>,
tun: Arc<Tun>,
incoming: AsyncMutex<Receiver<Bytes>>,
local_addr: SocketAddrV4,
remote_addr: SocketAddrV4,
incoming: flume::Receiver<Bytes>,
local_addr: SocketAddr,
remote_addr: SocketAddr,
seq: AtomicU32,
ack: AtomicU32,
last_ack: AtomicU32,
state: State,
}
@@ -120,22 +123,23 @@ impl Socket {
fn new(
shared: Arc<Shared>,
tun: Arc<Tun>,
local_addr: SocketAddrV4,
remote_addr: SocketAddrV4,
local_addr: SocketAddr,
remote_addr: SocketAddr,
ack: Option<u32>,
state: State,
) -> (Socket, Sender<Bytes>) {
let (incoming_tx, incoming_rx) = mpsc::channel(MPSC_BUFFER_LEN);
) -> (Socket, flume::Sender<Bytes>) {
let (incoming_tx, incoming_rx) = flume::bounded(MPMC_BUFFER_LEN);
(
Socket {
shared,
tun,
incoming: AsyncMutex::new(incoming_rx),
incoming: incoming_rx,
local_addr,
remote_addr,
seq: AtomicU32::new(0),
ack: AtomicU32::new(ack.unwrap_or(0)),
last_ack: AtomicU32::new(ack.unwrap_or(0)),
state,
},
incoming_tx,
@@ -143,11 +147,14 @@ impl Socket {
}
fn build_tcp_packet(&self, flags: u16, payload: Option<&[u8]>) -> Bytes {
let ack = self.ack.load(Ordering::Relaxed);
self.last_ack.store(ack, Ordering::Relaxed);
build_tcp_packet(
self.local_addr,
self.remote_addr,
self.seq.load(Ordering::Relaxed),
self.ack.load(Ordering::Relaxed),
ack,
flags,
payload,
)
@@ -165,12 +172,7 @@ impl Socket {
State::Established => {
let buf = self.build_tcp_packet(tcp::TcpFlags::ACK, Some(payload));
self.seq.fetch_add(payload.len() as u32, Ordering::Relaxed);
tokio::select! {
res = self.tun.send(&buf) => {
res.ok().and(Some(()))
},
}
self.tun.send(&buf).await.ok().and(Some(()))
}
_ => unreachable!(),
}
@@ -186,9 +188,8 @@ impl Socket {
pub async fn recv(&self, buf: &mut [u8]) -> Option<usize> {
match self.state {
State::Established => {
let mut incoming = self.incoming.lock().await;
incoming.recv().await.and_then(|raw_buf| {
let (_v4_packet, tcp_packet) = parse_ipv4_packet(&raw_buf);
self.incoming.recv_async().await.ok().and_then(|raw_buf| {
let (_v4_packet, tcp_packet) = parse_ip_packet(&raw_buf).unwrap();
if (tcp_packet.get_flags() & tcp::TcpFlags::RST) != 0 {
info!("Connection {} reset by peer", self);
@@ -197,8 +198,18 @@ impl Socket {
let payload = tcp_packet.payload();
self.ack
.store(tcp_packet.get_sequence().wrapping_add(1), Ordering::Relaxed);
let new_ack = tcp_packet.get_sequence().wrapping_add(payload.len() as u32);
let last_ask = self.last_ack.load(Ordering::Relaxed);
self.ack.store(new_ack, Ordering::Relaxed);
if new_ack.overflowing_sub(last_ask).0 > MAX_UNACKED_LEN {
let buf = self.build_tcp_packet(tcp::TcpFlags::ACK, None);
if let Err(e) = self.tun.try_send(&buf) {
// This should not really happen as we have not sent anything for
// quite some time...
info!("Connection {} unable to send idling ACK back: {}", self, e)
}
}
buf[..payload.len()].copy_from_slice(payload);
@@ -220,10 +231,10 @@ impl Socket {
info!("Sent SYN + ACK to client");
}
State::SynReceived => {
let res = time::timeout(TIMEOUT, self.incoming.lock().await.recv()).await;
let res = time::timeout(TIMEOUT, self.incoming.recv_async()).await;
if let Ok(buf) = res {
let buf = buf.unwrap();
let (_v4_packet, tcp_packet) = parse_ipv4_packet(&buf);
let (_v4_packet, tcp_packet) = parse_ip_packet(&buf).unwrap();
if (tcp_packet.get_flags() & tcp::TcpFlags::RST) != 0 {
return;
@@ -264,10 +275,10 @@ impl Socket {
info!("Sent SYN to server");
}
State::SynSent => {
match time::timeout(TIMEOUT, self.incoming.lock().await.recv()).await {
match time::timeout(TIMEOUT, self.incoming.recv_async()).await {
Ok(buf) => {
let buf = buf.unwrap();
let (_v4_packet, tcp_packet) = parse_ipv4_packet(&buf);
let (_v4_packet, tcp_packet) = parse_ip_packet(&buf).unwrap();
if (tcp_packet.get_flags() & tcp::TcpFlags::RST) != 0 {
return None;
@@ -315,7 +326,14 @@ impl Drop for Socket {
// purge cache
self.shared.tuples_purge.send(tuple).unwrap();
let buf = self.build_tcp_packet(tcp::TcpFlags::RST, None);
let buf = build_tcp_packet(
self.local_addr,
self.remote_addr,
self.seq.load(Ordering::Relaxed),
0,
tcp::TcpFlags::RST,
None,
);
if let Err(e) = self.tun.try_send(&buf) {
warn!("Unable to send RST to remote end: {}", e);
}
@@ -341,7 +359,7 @@ impl Stack {
/// When more than one [`Tun`](tokio_tun::Tun) object is passed in, same amount
/// of reader will be spawned later. This allows user to utilize the performance
/// benefit of Multiqueue Tun support on machines with SMP.
pub fn new(tun: Vec<Tun>) -> Stack {
pub fn new(tun: Vec<Tun>, local_ip: Ipv4Addr, local_ip6: Option<Ipv6Addr>) -> Stack {
let tun: Vec<Arc<Tun>> = tun.into_iter().map(Arc::new).collect();
let (ready_tx, ready_rx) = mpsc::channel(MPSC_BUFFER_LEN);
let (tuples_purge_tx, _tuples_purge_rx) = broadcast::channel(16);
@@ -352,7 +370,6 @@ impl Stack {
ready: ready_tx,
tuples_purge: tuples_purge_tx.clone(),
});
let local_ip = tun[0].destination().unwrap();
for t in tun {
tokio::spawn(Stack::reader_task(
@@ -365,6 +382,7 @@ impl Stack {
Stack {
shared,
local_ip,
local_ip6,
ready: ready_rx,
}
}
@@ -381,10 +399,17 @@ impl Stack {
/// Connects to the remote end. `None` returned means
/// the connection attempt failed.
pub async fn connect(&mut self, addr: SocketAddrV4) -> Option<Socket> {
pub async fn connect(&mut self, addr: SocketAddr) -> Option<Socket> {
let mut rng = SmallRng::from_entropy();
let local_port: u16 = rng.gen_range(1024..65535);
let local_addr = SocketAddrV4::new(self.local_ip, local_port);
let local_addr = SocketAddr::new(
if addr.is_ipv4() {
IpAddr::V4(self.local_ip)
} else {
IpAddr::V6(self.local_ip6.expect("IPv6 local address undefined"))
},
local_port,
);
let tuple = AddrTuple::new(local_addr, addr);
let (mut sock, incoming) = Socket::new(
self.shared.clone(),
@@ -408,7 +433,7 @@ impl Stack {
shared: Arc<Shared>,
mut tuples_purge: broadcast::Receiver<AddrTuple>,
) {
let mut tuples: HashMap<AddrTuple, Sender<Bytes>> = HashMap::new();
let mut tuples: HashMap<AddrTuple, flume::Sender<Bytes>> = HashMap::new();
loop {
let mut buf = BytesMut::with_capacity(MAX_PACKET_LEN);
@@ -420,95 +445,95 @@ impl Stack {
buf.truncate(size);
let buf = buf.freeze();
if buf[0] >> 4 != 4 {
// not an IPv4 packet
continue;
}
match parse_ip_packet(&buf) {
Some((ip_packet, tcp_packet)) => {
let local_addr =
SocketAddr::new(ip_packet.get_destination(), tcp_packet.get_destination());
let remote_addr = SocketAddr::new(ip_packet.get_source(), tcp_packet.get_source());
let (ip_packet, tcp_packet) = parse_ipv4_packet(&buf);
let local_addr =
SocketAddrV4::new(ip_packet.get_destination(), tcp_packet.get_destination());
let remote_addr = SocketAddrV4::new(ip_packet.get_source(), tcp_packet.get_source());
let tuple = AddrTuple::new(local_addr, remote_addr);
if let Some(c) = tuples.get(&tuple) {
if c.send_async(buf).await.is_err() {
trace!("Cache hit, but receiver already closed, dropping packet");
}
let tuple = AddrTuple::new(local_addr, remote_addr);
if let Some(c) = tuples.get(&tuple) {
if c.send(buf).await.is_err() {
trace!("Cache hit, but receiver already closed, dropping packet");
continue;
// If not Ok, receiver has been closed and just fall through to the slow
// path below
} else {
trace!("Cache miss, checking the shared tuples table for connection");
let sender = {
let tuples = shared.tuples.read().unwrap();
tuples.get(&tuple).cloned()
};
if let Some(c) = sender {
trace!("Storing connection information into local tuples");
tuples.insert(tuple, c.clone());
c.send_async(buf).await.unwrap();
continue;
}
}
if tcp_packet.get_flags() == tcp::TcpFlags::SYN
&& shared
.listening
.read()
.unwrap()
.contains(&tcp_packet.get_destination())
{
// SYN seen on listening socket
if tcp_packet.get_sequence() == 0 {
let (sock, incoming) = Socket::new(
shared.clone(),
tun.clone(),
local_addr,
remote_addr,
Some(tcp_packet.get_sequence() + 1),
State::Idle,
);
assert!(shared
.tuples
.write()
.unwrap()
.insert(tuple, incoming)
.is_none());
tokio::spawn(sock.accept());
} else {
trace!("Bad TCP SYN packet from {}, sending RST", remote_addr);
let buf = build_tcp_packet(
local_addr,
remote_addr,
0,
tcp_packet.get_sequence() + tcp_packet.payload().len() as u32 + 1, // +1 because of SYN flag set
tcp::TcpFlags::RST | tcp::TcpFlags::ACK,
None,
);
shared.tun[0].try_send(&buf).unwrap();
}
} else if (tcp_packet.get_flags() & tcp::TcpFlags::RST) == 0 {
info!("Unknown TCP packet from {}, sending RST", remote_addr);
let buf = build_tcp_packet(
local_addr,
remote_addr,
tcp_packet.get_acknowledgement(),
tcp_packet.get_sequence() + tcp_packet.payload().len() as u32,
tcp::TcpFlags::RST | tcp::TcpFlags::ACK,
None,
);
shared.tun[0].try_send(&buf).unwrap();
}
}
continue;
// If not Ok, receiver has been closed and just fall through to the slow
// path below
} else {
trace!("Cache miss, checking the shared tuples table for connection");
let sender = {
let tuples = shared.tuples.read().unwrap();
tuples.get(&tuple).cloned()
};
if let Some(c) = sender {
trace!("Storing connection information into local tuples");
tuples.insert(tuple, c.clone());
c.send(buf).await.unwrap();
None => {
continue;
}
}
if tcp_packet.get_flags() == tcp::TcpFlags::SYN
&& shared
.listening
.read()
.unwrap()
.contains(&tcp_packet.get_destination())
{
// SYN seen on listening socket
if tcp_packet.get_sequence() == 0 {
let (sock, incoming) = Socket::new(
shared.clone(),
tun.clone(),
local_addr,
remote_addr,
Some(tcp_packet.get_sequence() + 1),
State::Idle,
);
assert!(shared
.tuples
.write()
.unwrap()
.insert(tuple, incoming)
.is_none());
tokio::spawn(sock.accept());
} else {
trace!("Bad TCP SYN packet from {}, sending RST", remote_addr);
let buf = build_tcp_packet(
local_addr,
remote_addr,
0,
tcp_packet.get_sequence() + 1,
tcp::TcpFlags::RST,
None,
);
shared.tun[0].try_send(&buf).unwrap();
}
} else if (tcp_packet.get_flags() & tcp::TcpFlags::RST) == 0 {
info!("Unknown TCP packet from {}, sending RST", remote_addr);
let buf = build_tcp_packet(
local_addr,
remote_addr,
tcp_packet.get_acknowledgement(),
0,
tcp::TcpFlags::RST,
None,
);
shared.tun[0].try_send(&buf).unwrap();
}
},
tuple = tuples_purge.recv() => {
let tuple = tuple.unwrap();
tuples.remove(&tuple);
trace!("Removed cached tuple");
trace!("Removed cached tuple: {:?}", tuple);
}
}
}

View File

@@ -1,45 +1,85 @@
use bytes::{Bytes, BytesMut};
use internet_checksum::Checksum;
use pnet::packet::Packet;
use pnet::packet::{ip, ipv4, tcp};
use pnet::packet::{ip, ipv4, ipv6, tcp};
use std::convert::TryInto;
use std::net::SocketAddrV4;
use std::net::{IpAddr, SocketAddr};
const IPV4_HEADER_LEN: usize = 20;
const IPV6_HEADER_LEN: usize = 40;
const TCP_HEADER_LEN: usize = 20;
pub const MAX_PACKET_LEN: usize = 1500;
pub enum IPPacket<'p> {
V4(ipv4::Ipv4Packet<'p>),
V6(ipv6::Ipv6Packet<'p>),
}
impl<'a> IPPacket<'a> {
pub fn get_source(&self) -> IpAddr {
match self {
IPPacket::V4(p) => IpAddr::V4(p.get_source()),
IPPacket::V6(p) => IpAddr::V6(p.get_source()),
}
}
pub fn get_destination(&self) -> IpAddr {
match self {
IPPacket::V4(p) => IpAddr::V4(p.get_destination()),
IPPacket::V6(p) => IpAddr::V6(p.get_destination()),
}
}
}
pub fn build_tcp_packet(
local_addr: SocketAddrV4,
remote_addr: SocketAddrV4,
local_addr: SocketAddr,
remote_addr: SocketAddr,
seq: u32,
ack: u32,
flags: u16,
payload: Option<&[u8]>,
) -> Bytes {
let ip_header_len = match local_addr {
SocketAddr::V4(_) => IPV4_HEADER_LEN,
SocketAddr::V6(_) => IPV6_HEADER_LEN,
};
let wscale = (flags & tcp::TcpFlags::SYN) != 0;
let tcp_header_len = TCP_HEADER_LEN + if wscale { 4 } else { 0 }; // nop + wscale
let tcp_total_len = tcp_header_len + payload.map_or(0, |payload| payload.len());
let total_len = IPV4_HEADER_LEN + tcp_total_len;
let total_len = ip_header_len + tcp_total_len;
let mut buf = BytesMut::with_capacity(total_len);
buf.resize(total_len, 0);
let mut v4_buf = buf.split_to(IPV4_HEADER_LEN);
let mut ip_buf = buf.split_to(ip_header_len);
let mut tcp_buf = buf.split_to(tcp_total_len);
assert_eq!(0, buf.len());
let mut v4 = ipv4::MutableIpv4Packet::new(&mut v4_buf).unwrap();
v4.set_version(4);
v4.set_header_length(IPV4_HEADER_LEN as u8 / 4);
v4.set_next_level_protocol(ip::IpNextHeaderProtocols::Tcp);
v4.set_ttl(64);
v4.set_source(*local_addr.ip());
v4.set_destination(*remote_addr.ip());
v4.set_total_length(total_len.try_into().unwrap());
v4.set_flags(ipv4::Ipv4Flags::DontFragment);
let mut cksm = Checksum::new();
cksm.add_bytes(v4.packet());
v4.set_checksum(u16::from_be_bytes(cksm.checksum()));
match (local_addr, remote_addr) {
(SocketAddr::V4(local), SocketAddr::V4(remote)) => {
let mut v4 = ipv4::MutableIpv4Packet::new(&mut ip_buf).unwrap();
v4.set_version(4);
v4.set_header_length(IPV4_HEADER_LEN as u8 / 4);
v4.set_next_level_protocol(ip::IpNextHeaderProtocols::Tcp);
v4.set_ttl(64);
v4.set_source(*local.ip());
v4.set_destination(*remote.ip());
v4.set_total_length(total_len.try_into().unwrap());
v4.set_flags(ipv4::Ipv4Flags::DontFragment);
let mut cksm = Checksum::new();
cksm.add_bytes(v4.packet());
v4.set_checksum(u16::from_be_bytes(cksm.checksum()));
}
(SocketAddr::V6(local), SocketAddr::V6(remote)) => {
let mut v6 = ipv6::MutableIpv6Packet::new(&mut ip_buf).unwrap();
v6.set_version(6);
v6.set_payload_length(tcp_total_len.try_into().unwrap());
v6.set_next_header(ip::IpNextHeaderProtocols::Tcp);
v6.set_hop_limit(64);
v6.set_source(*local.ip());
v6.set_destination(*remote.ip());
}
_ => unreachable!(),
};
let mut tcp = tcp::MutableTcpPacket::new(&mut tcp_buf).unwrap();
tcp.set_window(0xffff);
@@ -59,24 +99,55 @@ pub fn build_tcp_packet(
}
let mut cksm = Checksum::new();
cksm.add_bytes(&local_addr.ip().octets());
cksm.add_bytes(&remote_addr.ip().octets());
let ip::IpNextHeaderProtocol(tcp_protocol) = ip::IpNextHeaderProtocols::Tcp;
let mut pseudo = [0u8, tcp_protocol, 0, 0];
pseudo[2..].copy_from_slice(&(tcp_total_len as u16).to_be_bytes());
cksm.add_bytes(&pseudo);
match (local_addr, remote_addr) {
(SocketAddr::V4(local), SocketAddr::V4(remote)) => {
cksm.add_bytes(&local.ip().octets());
cksm.add_bytes(&remote.ip().octets());
let mut pseudo = [0u8, tcp_protocol, 0, 0];
pseudo[2..].copy_from_slice(&(tcp_total_len as u16).to_be_bytes());
cksm.add_bytes(&pseudo);
}
(SocketAddr::V6(local), SocketAddr::V6(remote)) => {
cksm.add_bytes(&local.ip().octets());
cksm.add_bytes(&remote.ip().octets());
let mut pseudo = [0u8, 0, 0, 0, 0, 0, 0, tcp_protocol];
pseudo[0..4].copy_from_slice(&(tcp_total_len as u32).to_be_bytes());
cksm.add_bytes(&pseudo);
}
_ => unreachable!(),
};
cksm.add_bytes(tcp.packet());
tcp.set_checksum(u16::from_be_bytes(cksm.checksum()));
v4_buf.unsplit(tcp_buf);
v4_buf.freeze()
ip_buf.unsplit(tcp_buf);
ip_buf.freeze()
}
pub fn parse_ipv4_packet(buf: &Bytes) -> (ipv4::Ipv4Packet, tcp::TcpPacket) {
let v4 = ipv4::Ipv4Packet::new(buf).unwrap();
let tcp = tcp::TcpPacket::new(&buf[IPV4_HEADER_LEN..]).unwrap();
pub fn parse_ip_packet(buf: &Bytes) -> Option<(IPPacket, tcp::TcpPacket)> {
if buf[0] >> 4 == 4 {
let v4 = ipv4::Ipv4Packet::new(buf).unwrap();
if v4.get_next_level_protocol() != ip::IpNextHeaderProtocols::Tcp {
return None;
}
(v4, tcp)
let tcp = tcp::TcpPacket::new(&buf[IPV4_HEADER_LEN..]).unwrap();
Some((IPPacket::V4(v4), tcp))
} else if buf[0] >> 4 == 6 {
let v6 = ipv6::Ipv6Packet::new(buf).unwrap();
if v6.get_next_header() != ip::IpNextHeaderProtocols::Tcp {
return None;
}
let tcp = tcp::TcpPacket::new(&buf[IPV6_HEADER_LEN..]).unwrap();
Some((IPPacket::V6(v6), tcp))
} else {
None
}
}
#[cfg(all(test, feature = "benchmark"))]

BIN
images/packet-headers.png Normal file

Binary file not shown.

After

Width:  |  Height:  |  Size: 30 KiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 15 KiB

View File

@@ -1,6 +1,6 @@
[package]
name = "phantun"
version = "0.3.0"
version = "0.4.0"
edition = "2021"
authors = ["Datong Sun <dndx@idndx.com>"]
license = "MIT OR Apache-2.0"
@@ -13,10 +13,12 @@ Layer 3 & Layer 4 (NAPT) firewalls/NATs.
[dependencies]
clap = { version = "3.0", features = ["cargo"] }
socket2 = { version = "0.4", features = ["all"] }
fake-tcp = { path = "../fake-tcp", version = "0.2" }
fake-tcp = { path = "../fake-tcp", version = "0.4" }
tokio = { version = "1.14", features = ["full"] }
tokio-util = "0.7"
log = "0.4"
pretty_env_logger = "0.4"
tokio-tun = "0.5"
num_cpus = "1.13"
neli = "0.6"
nix = "0.23"

View File

@@ -2,8 +2,9 @@ use clap::{crate_version, Arg, Command};
use fake_tcp::packet::MAX_PACKET_LEN;
use fake_tcp::{Socket, Stack};
use log::{debug, error, info};
use phantun::utils::new_udp_reuseport;
use phantun::utils::{assign_ipv6_address, new_udp_reuseport};
use std::collections::HashMap;
use std::io;
use std::net::{Ipv4Addr, SocketAddr};
use std::sync::Arc;
use tokio::sync::{Notify, RwLock};
@@ -14,7 +15,7 @@ use tokio_util::sync::CancellationToken;
use phantun::UDP_TTL;
#[tokio::main]
async fn main() {
async fn main() -> io::Result<()> {
pretty_env_logger::init();
let matches = Command::new("Phantun Client")
@@ -35,7 +36,7 @@ async fn main() {
.long("remote")
.required(true)
.value_name("IP or HOST NAME:PORT")
.help("Sets the address or host name and port where Phantun Client connects to Phantun Server")
.help("Sets the address or host name and port where Phantun Client connects to Phantun Server, IPv6 address need to be specified as: \"[IPv6]:PORT\"")
.takes_value(true),
)
.arg(
@@ -52,7 +53,7 @@ async fn main() {
.long("tun-local")
.required(false)
.value_name("IP")
.help("Sets the Tun interface local address (O/S's end)")
.help("Sets the Tun interface IPv4 local address (O/S's end)")
.default_value("192.168.200.1")
.takes_value(true),
)
@@ -61,12 +62,41 @@ async fn main() {
.long("tun-peer")
.required(false)
.value_name("IP")
.help("Sets the Tun interface destination (peer) address (Phantun Client's end). \
.help("Sets the Tun interface IPv4 destination (peer) address (Phantun Client's end). \
You will need to setup SNAT/MASQUERADE rules on your Internet facing interface \
in order for Phantun Client to connect to Phantun Server")
.default_value("192.168.200.2")
.takes_value(true),
)
.arg(
Arg::new("ipv4_only")
.long("ipv4-only")
.short('4')
.required(false)
.help("Do not assign IPv6 addresses to Tun interface")
.takes_value(false)
.conflicts_with_all(&["tun_local6", "tun_peer6"]),
)
.arg(
Arg::new("tun_local6")
.long("tun-local6")
.required(false)
.value_name("IP")
.help("Sets the Tun interface IPv6 local address (O/S's end)")
.default_value("fec8::1")
.takes_value(true),
)
.arg(
Arg::new("tun_peer6")
.long("tun-peer6")
.required(false)
.value_name("IP")
.help("Sets the Tun interface IPv6 destination (peer) address (Phantun Client's end). \
You will need to setup SNAT/MASQUERADE rules on your Internet facing interface \
in order for Phantun Client to connect to Phantun Server")
.default_value("fec8::2")
.takes_value(true),
)
.get_matches();
let local_addr: SocketAddr = matches
@@ -75,16 +105,13 @@ async fn main() {
.parse()
.expect("bad local address");
let ipv4_only = matches.is_present("ipv4_only");
let remote_addr = tokio::net::lookup_host(matches.value_of("remote").unwrap())
.await
.expect("bad remote address or host")
.find(|addr| addr.is_ipv4())
.expect("unable to resolve remote host name or no valid A record was returned");
let remote_addr = if let SocketAddr::V4(addr) = remote_addr {
addr
} else {
unreachable!();
};
.find(|addr| !ipv4_only || addr.is_ipv4())
.expect("unable to resolve remote host name");
info!("Remote address is: {}", remote_addr);
let tun_local: Ipv4Addr = matches
@@ -98,10 +125,26 @@ async fn main() {
.parse()
.expect("bad peer address for Tun interface");
let (tun_local6, tun_peer6) = if ipv4_only {
(None, None)
} else {
(
matches
.value_of("tun_local6")
.map(|v| v.parse().expect("bad local address for Tun interface")),
matches
.value_of("tun_peer6")
.map(|v| v.parse().expect("bad peer address for Tun interface")),
)
};
let tun_name = matches.value_of("tun").unwrap();
let num_cpus = num_cpus::get();
info!("{} cores available", num_cpus);
let tun = TunBuilder::new()
.name(matches.value_of("tun").unwrap()) // if name is empty, then it is set by kernel.
.name(tun_name) // if name is empty, then it is set by kernel.
.tap(false) // false (default): TUN, true: TAP.
.packet_info(false) // false: IFF_NO_PI, default is true.
.up() // or set it up manually using `sudo ip link set <tun-name> up`.
@@ -110,130 +153,135 @@ async fn main() {
.try_build_mq(num_cpus)
.unwrap();
if remote_addr.is_ipv6() {
assign_ipv6_address(tun[0].name(), tun_local6.unwrap(), tun_peer6.unwrap());
}
info!("Created TUN device {}", tun[0].name());
let udp_sock = Arc::new(new_udp_reuseport(local_addr));
let connections = Arc::new(RwLock::new(HashMap::<SocketAddr, Arc<Socket>>::new()));
let mut stack = Stack::new(tun);
let mut stack = Stack::new(tun, tun_peer, tun_peer6);
let main_loop = tokio::spawn(async move {
let mut buf_r = [0u8; MAX_PACKET_LEN];
loop {
tokio::select! {
Ok((size, addr)) = udp_sock.recv_from(&mut buf_r) => {
// seen UDP packet to listening socket, this means:
// 1. It is a new UDP connection, or
// 2. It is some extra packets not filtered by more specific
// connected UDP socket yet
if let Some(sock) = connections.read().await.get(&addr) {
sock.send(&buf_r[..size]).await;
continue;
}
let (size, addr) = udp_sock.recv_from(&mut buf_r).await?;
// seen UDP packet to listening socket, this means:
// 1. It is a new UDP connection, or
// 2. It is some extra packets not filtered by more specific
// connected UDP socket yet
if let Some(sock) = connections.read().await.get(&addr) {
sock.send(&buf_r[..size]).await;
continue;
}
info!("New UDP client from {}", addr);
let sock = stack.connect(remote_addr).await;
if sock.is_none() {
error!("Unable to connect to remote {}", remote_addr);
continue;
}
info!("New UDP client from {}", addr);
let sock = stack.connect(remote_addr).await;
if sock.is_none() {
error!("Unable to connect to remote {}", remote_addr);
continue;
}
let sock = Arc::new(sock.unwrap());
// send first packet
let res = sock.send(&buf_r[..size]).await;
if res.is_none() {
continue;
}
let sock = Arc::new(sock.unwrap());
// send first packet
let res = sock.send(&buf_r[..size]).await;
if res.is_none() {
continue;
}
assert!(connections.write().await.insert(addr, sock.clone()).is_none());
debug!("inserted fake TCP socket into connection table");
assert!(connections
.write()
.await
.insert(addr, sock.clone())
.is_none());
debug!("inserted fake TCP socket into connection table");
// spawn "fastpath" UDP socket and task, this will offload main task
// from forwarding UDP packets
// spawn "fastpath" UDP socket and task, this will offload main task
// from forwarding UDP packets
let packet_received = Arc::new(Notify::new());
let quit = CancellationToken::new();
let packet_received = Arc::new(Notify::new());
let quit = CancellationToken::new();
for i in 0..num_cpus {
let sock = sock.clone();
let quit = quit.child_token();
let packet_received = packet_received.clone();
for i in 0..num_cpus {
let sock = sock.clone();
let quit = quit.clone();
let packet_received = packet_received.clone();
tokio::spawn(async move {
let mut buf_udp = [0u8; MAX_PACKET_LEN];
let mut buf_tcp = [0u8; MAX_PACKET_LEN];
let udp_sock = new_udp_reuseport(local_addr);
udp_sock.connect(addr).await.unwrap();
tokio::spawn(async move {
let mut buf_udp = [0u8; MAX_PACKET_LEN];
let mut buf_tcp = [0u8; MAX_PACKET_LEN];
let udp_sock = new_udp_reuseport(local_addr);
udp_sock.connect(addr).await.unwrap();
loop {
tokio::select! {
Ok(size) = udp_sock.recv(&mut buf_udp) => {
if sock.send(&buf_udp[..size]).await.is_none() {
debug!("removed fake TCP socket from connections table");
quit.cancel();
return;
}
packet_received.notify_one();
},
res = sock.recv(&mut buf_tcp) => {
match res {
Some(size) => {
if size > 0 {
if let Err(e) = udp_sock.send(&buf_tcp[..size]).await {
error!("Unable to send UDP packet to {}: {}, closing connection", e, addr);
quit.cancel();
return;
}
}
},
None => {
debug!("removed fake TCP socket from connections table");
quit.cancel();
return;
},
}
packet_received.notify_one();
},
_ = quit.cancelled() => {
debug!("worker {} terminated", i);
return;
},
};
}
});
}
let connections = connections.clone();
tokio::spawn(async move {
loop {
let read_timeout = time::sleep(UDP_TTL);
let packet_received_fut = packet_received.notified();
tokio::select! {
_ = read_timeout => {
info!("No traffic seen in the last {:?}, closing connection", UDP_TTL);
connections.write().await.remove(&addr);
loop {
tokio::select! {
Ok(size) = udp_sock.recv(&mut buf_udp) => {
if sock.send(&buf_udp[..size]).await.is_none() {
debug!("removed fake TCP socket from connections table");
quit.cancel();
return;
},
_ = quit.cancelled() => {
connections.write().await.remove(&addr);
debug!("removed fake TCP socket from connections table");
return;
},
_ = packet_received_fut => {},
}
}
});
},
}
packet_received.notify_one();
},
res = sock.recv(&mut buf_tcp) => {
match res {
Some(size) => {
if size > 0 {
if let Err(e) = udp_sock.send(&buf_tcp[..size]).await {
error!("Unable to send UDP packet to {}: {}, closing connection", e, addr);
quit.cancel();
return;
}
}
},
None => {
debug!("removed fake TCP socket from connections table");
quit.cancel();
return;
},
}
packet_received.notify_one();
},
_ = quit.cancelled() => {
debug!("worker {} terminated", i);
return;
},
};
}
});
}
let connections = connections.clone();
tokio::spawn(async move {
loop {
let read_timeout = time::sleep(UDP_TTL);
let packet_received_fut = packet_received.notified();
tokio::select! {
_ = read_timeout => {
info!("No traffic seen in the last {:?}, closing connection", UDP_TTL);
connections.write().await.remove(&addr);
debug!("removed fake TCP socket from connections table");
quit.cancel();
return;
},
_ = quit.cancelled() => {
connections.write().await.remove(&addr);
debug!("removed fake TCP socket from connections table");
return;
},
_ = packet_received_fut => {},
}
}
});
}
});
tokio::join!(main_loop).0.unwrap();
tokio::join!(main_loop).0.unwrap()
}

View File

@@ -2,7 +2,8 @@ use clap::{crate_version, Arg, Command};
use fake_tcp::packet::MAX_PACKET_LEN;
use fake_tcp::Stack;
use log::{debug, error, info};
use phantun::utils::new_udp_reuseport;
use phantun::utils::{assign_ipv6_address, new_udp_reuseport};
use std::io;
use std::net::Ipv4Addr;
use std::sync::Arc;
use tokio::net::UdpSocket;
@@ -14,7 +15,7 @@ use tokio_util::sync::CancellationToken;
use phantun::UDP_TTL;
#[tokio::main]
async fn main() {
async fn main() -> io::Result<()> {
pretty_env_logger::init();
let matches = Command::new("Phantun Server")
@@ -67,6 +68,35 @@ async fn main() {
.default_value("192.168.201.2")
.takes_value(true),
)
.arg(
Arg::new("ipv4_only")
.long("ipv4-only")
.short('4')
.required(false)
.help("Do not assign IPv6 addresses to Tun interface")
.takes_value(false)
.conflicts_with_all(&["tun_local6", "tun_peer6"]),
)
.arg(
Arg::new("tun_local6")
.long("tun-local6")
.required(false)
.value_name("IP")
.help("Sets the Tun interface IPv6 local address (O/S's end)")
.default_value("fec9::1")
.takes_value(true),
)
.arg(
Arg::new("tun_peer6")
.long("tun-peer6")
.required(false)
.value_name("IP")
.help("Sets the Tun interface IPv6 destination (peer) address (Phantun Client's end). \
You will need to setup SNAT/MASQUERADE rules on your Internet facing interface \
in order for Phantun Client to connect to Phantun Server")
.default_value("fec9::2")
.takes_value(true),
)
.get_matches();
let local_port: u16 = matches
@@ -93,10 +123,26 @@ async fn main() {
.parse()
.expect("bad peer address for Tun interface");
let (tun_local6, tun_peer6) = if matches.is_present("ipv4_only") {
(None, None)
} else {
(
matches
.value_of("tun_local6")
.map(|v| v.parse().expect("bad local address for Tun interface")),
matches
.value_of("tun_peer6")
.map(|v| v.parse().expect("bad peer address for Tun interface")),
)
};
let tun_name = matches.value_of("tun").unwrap();
let num_cpus = num_cpus::get();
info!("{} cores available", num_cpus);
let tun = TunBuilder::new()
.name(matches.value_of("tun").unwrap()) // if name is empty, then it is set by kernel.
.name(tun_name) // if name is empty, then it is set by kernel.
.tap(false) // false (default): TUN, true: TAP.
.packet_info(false) // false: IFF_NO_PI, default is true.
.up() // or set it up manually using `sudo ip link set <tun-name> up`.
@@ -105,10 +151,14 @@ async fn main() {
.try_build_mq(num_cpus)
.unwrap();
if let (Some(tun_local6), Some(tun_peer6)) = (tun_local6, tun_peer6) {
assign_ipv6_address(tun[0].name(), tun_local6, tun_peer6);
}
info!("Created TUN device {}", tun[0].name());
//thread::sleep(time::Duration::from_secs(5));
let mut stack = Stack::new(tun);
let mut stack = Stack::new(tun, tun_local, tun_local6);
stack.listen(local_port);
info!("Listening on {}", local_port);
@@ -127,14 +177,13 @@ async fn main() {
} else {
"[::]:0"
})
.await
.unwrap();
let local_addr = udp_sock.local_addr().unwrap();
.await?;
let local_addr = udp_sock.local_addr()?;
drop(udp_sock);
for i in 0..num_cpus {
let sock = sock.clone();
let quit = quit.child_token();
let quit = quit.clone();
let packet_received = packet_received.clone();
let udp_sock = new_udp_reuseport(local_addr);
@@ -198,5 +247,5 @@ async fn main() {
}
});
tokio::join!(main_loop).0.unwrap();
tokio::join!(main_loop).0.unwrap()
}

View File

@@ -1,4 +1,15 @@
use std::net::SocketAddr;
use neli::{
consts::{
nl::{NlmF, NlmFFlags},
rtnl::{Ifa, IfaFFlags, RtAddrFamily, Rtm},
socket::NlFamily,
},
nl::{NlPayload, Nlmsghdr},
rtnl::{Ifaddrmsg, Rtattr},
socket::NlSocketHandle,
types::RtBuffer,
};
use std::net::{Ipv6Addr, SocketAddr};
use tokio::net::UdpSocket;
pub fn new_udp_reuseport(local_addr: SocketAddr) -> UdpSocket {
@@ -20,3 +31,30 @@ pub fn new_udp_reuseport(local_addr: SocketAddr) -> UdpSocket {
let udp_sock: std::net::UdpSocket = udp_sock.into();
udp_sock.try_into().unwrap()
}
pub fn assign_ipv6_address(device_name: &str, local: Ipv6Addr, peer: Ipv6Addr) {
let index = nix::net::if_::if_nametoindex(device_name).unwrap();
let mut rtnl = NlSocketHandle::connect(NlFamily::Route, None, &[]).unwrap();
let mut rtattrs = RtBuffer::new();
rtattrs.push(Rtattr::new(None, Ifa::Local, &local.octets()[..]).unwrap());
rtattrs.push(Rtattr::new(None, Ifa::Address, &peer.octets()[..]).unwrap());
let ifaddrmsg = Ifaddrmsg {
ifa_family: RtAddrFamily::Inet6,
ifa_prefixlen: 128,
ifa_flags: IfaFFlags::empty(),
ifa_scope: 0,
ifa_index: index as i32,
rtattrs,
};
let nl_header = Nlmsghdr::new(
None,
Rtm::Newaddr,
NlmFFlags::new(&[NlmF::Request]),
None,
None,
NlPayload::Payload(ifaddrmsg),
);
rtnl.send(nl_header).unwrap();
}