magic-wormhole-transit-relay/docs/transit.md

= Transit Protocol =

The Transit protocol is responsible for establishing an encrypted
bidirectional record stream between two programs. It must be given a "transit
key" and a set of "hints" which help locate the other end (which are both
delivered by Wormhole).

The protocol tries hard to create a **direct** connection between the two
ends, but if that fails, it uses a centralized relay server to ferry data
between two separate TCP streams (one to each client).

The current implementation starts with the following:

* detect all of the host's IP addresses
* listen on a random TCP port
* offers the (address,port) pairs as hints

The other side will attempt to connect to each of those ports, as well as
listening on its own socket. After a few seconds without success, they will
both connect to a relay server.

== Roles ==

The Transit protocol has pre-defined "Sender" and "Receiver" roles (unlike
Wormhole, which is symmetric/nobody-goes-first). Each connection must have
exactly one Sender and exactly one Receiver.

The connection itself is bidirectional: either side can send or receive
records. However the connection establishment mechanism needs to know who is
in charge, and the encryption layer needs a way to produce separate keys for
each side..

This may be relaxed in the future, much as Wormhole was.

== Records ==

Transit establishes a **record-pipe**, so the two sides can send and receive
whole records, rather than unframed bytes. This is a side-effect of the
encryption (which uses the NaCl "secretbox" function). The encryption adds 44
bytes of overhead to each record (4-byte length, 24-byte nonce, 32-byte MAC),
so you might want to use slightly larger records for efficiency. The maximum
record size is 2^32 bytes (4GiB). The whole record must be held in memory at
the same time, plus its ciphertext, so very large ciphertexts are not
recommended.

Transit provides **confidentiality**, **integrity**, and **ordering** of
records. Passive attackers can only do the following:

* learn the size and transmission time of each record
* learn the sending and destination IP addresses

In addition, an active attacker is able to:

* delay delivery of individual records, while maintaining ordering (if they
  delay record #4, they must delay #5 and later as well)
* terminate the connection at any time

If either side receives a corrupted or out-of-order record, they drop the
connection. Attackers cannot modify the contents of a record, or change the
order of the records, without being detected and the connection being
dropped. If a record is lost (e.g. the receiver observers records #1,#2,#4,
but not #3), the connection is dropped when the unexpected sequence number is
received.

== Handshake ==

The transit key is used to derive several secondary keys. Two of them are
used as a "handshake", to distinguish correct Transit connections from other
programs that happen to connect to the Transit sockets by mistake or malice.

The handshake is also responsible for choosing exactly one TCP connection to
use, even though multiple outbound and inbound connections are being
attempted.

The SENDER-HANDSHAKE is the string `transit sender %s ready\n\n`, with the
`%s` replaced by a hex-encoded 32-byte HKDF derivative of the transit key,
using a "context string" of `transit_sender`. The RECEIVER-HANDSHAKE is the
same but with `receiver` instead of `sender` (both for the string and the
HKDF context).

The handshake protocol is like this:

* immediately upon socket connection being made, the Sender writes
  SENDER-HANDSHAKE to the socket (regardless of whether the Sender initiated
  the TCP connection, or was listening on a socket and just accepted the
  connection)
* likewise the Receiver immediately writes RECEIVER-HANDSHAKE to either kind
  of socket
* if the Sender sees anything other than RECEIVER-HANDSHAKE as the first
  bytes on the wire, it hangs up
* likewise with the Receiver and SENDER-HANDSHAKE
* if the Sender sees that this is the first connection to get
  RECEIVER-HANDSHAKE, it sends `go\n`. If some other connection got there
  first, it hangs up (or sends `nevermind\n` and then hangs up, but this is
  mostly for debugging, and implementations should not depend upon it). After
  sending `go`, it switches to encrypted-record mode.
* if the Receiver sees `go\n`, it switches to encrypted-record mode. If the
  receiver sees anything else, or a disconnected socket, it disconnects.

To tolerate the inevitable race conditions created by multiple contending
sockets, only the Sender gets to decide which one wins: the first one to make
it past negotiation. Hopefully this is correlated with the fastest connection
pathway. The protocol ignores any socket that is not somewhat affiliated with
the matching Transit instance.

Hints will frequently point to local IP addresses (local to the other end)
which might be in use by unrelated nearby computers. The handshake helps to
ignore these spurious connections. It is still possible for an attacker to
cause the connection to fail, by intercepting both connections (to learn the
two handshakes), then making new connections to play back the recorded
handshakes, but this level of attacker could simply drop the user's packets
directly.

== Relay ==

The **Transit Relay** is a host which offers TURN-like services for
magic-wormhole instances. It uses a TCP-based protocol with a handshake to
determine which connection wants to be connected to which.

When connecting to a relay, the Transit client first writes RELAY-HANDSHAKE
to the socket, which is `please relay %s\n`, where `%s` is the hex-encoded
32-byte HKDF derivative of the transit key, using `transit_relay_token` as
the context. The client then waits for `ok\n`.

The relay waits for a second connection that uses the same token. When this
happens, the relay sends `ok\n` to both, then wires the connections together,
so that everything received after the token on one is written out (after the
ok) on the other. When either connection is lost, the other will be closed
(the relay does not support "half-close").

When clients use a relay connection, they perform the usual sender/receiver
handshake just after the `ok\n` is received: until that point they pretend
the connection doesn't even exist.

Direct connections are better, since they are faster and less expensive for
the relay operator. If there are any potentially-viable direct connection
hints available, the Transit instance will wait a few seconds before
attempting to use the relay. If it has no viable direct hints, it will start
using the relay right away. This prefers direct connections, but doesn't
introduce completely unnecessary stalls.

== API ==

First, create a Transit instance, giving it the connection information of the
transit relay. The application must know whether it should use a Sender or a
Receiver:

```python
from wormhole.blocking.transit import TransitSender
s = TransitSender("tcp:relayhost.example.org:12345")
```

Next, ask the Transit for its direct and relay hints. This should be
delivered to the other side via a Wormhole message (i.e. add them to a dict,
serialize it with JSON, send the result as a message with `wormhole.send()`).

```python
direct_hints = s.get_direct_hints()
relay_hints = s.get_relay_hints()
```

Then, perform the Wormhole exchange, which ought to give you the direct and
relay hints of the other side. Tell your Transit instance about their hints.

```python
s.add_their_direct_hints(their_direct_hints)
s.add_their_relay_hints(their_relay_hints)
```

Then use `wormhole.derive_key()` to obtain a shared key for Transit purposes,
and tell your Transit about it. Both sides must use the same derivation
string, and this string must not be used for any other purpose, but beyond
that it doesn't much matter what the exact string is.

```python
key = w.derive_key(application_id + "/transit-key")
s.set_transit_key(key)
```

Finally, tell the Transit instance to connect. This will yield a "record
pipe" object, on which records can be sent and received. If no connection can
be established within a timeout (defaults to 30 seconds), `connect()` will
throw an exception instead. The pipe can be closed with `close()`.

```python
rp = s.connect()
rp.send_record(b"my first record")
their_record = rp.receive_record()
rp.send_record(b"Greatest Hits)
other = rp.receive_record()
rp.close()
```

Records can be sent and received arbitrarily (you are not limited to taking
turns). However the blocking API does not provide a way to send records while
waiting for an inbound record. This *might* work with threads, but it has not
been tested.

== Twisted API ==

The same facilities are available in the asynchronous Twisted environment.
The difference is that some functions return Deferreds instead of immediate
values. The final record-pipe object is a Protocol (TBD: maybe this is a job
for Tubes?), which exposes `receive_record()` as a Deferred-returning
function that internally holds a queue of inbound records.

```python
from twisted.internet.defer import inlineCallbacks
from wormhole.twisted.transit import TransitSender
 
@inlineCallbacks
def do_transit():
    s = TransitSender(relay)
    my_relay_hints = s.get_relay_hints()
    my_direct_hints = yield s.get_direct_hints()
    # (send hints via wormhole)
    s.add_their_relay_hints(their_relay_hints)
    s.add_their_direct_hints(their_direct_hints)
    s.set_transit_key(key)
    rp = yield s.connect()
    rp.send_record(b"eponymous")
    them = yield rp.receive_record()
    yield rp.close()
```

This object also implements the `IConsumer`/`IProducer` protocols for
**bytes**, which means you can transfer a file by wiring up a file reader as
a Producer. Each chunk of bytes that the Producer generates will be put into
a single record. The Consumer interface works the same way. This enables
backpressure and flow-control: if the far end (or the network) cannot keep up
with the stream of data, the sender will wait for them to catch up before
filling buffers without bound.