Network Protocol

Date: 2009-01-07

Contents

Overview

The smart protocol provides a way to send a requests and corresponding responses to communicate with a remote bzr process.

Layering

Medium

At the bottom level there is either a socket, pipes, or an HTTP request/response. We call this layer the medium. It is responsible for carrying bytes between a client and server. For sockets, we have the idea that you have multiple requests and get a read error because the other side did shutdown. For pipes we have read pipe which will have a zero read which marks end-of-file. For HTTP server environment there is no end-of-stream because each request coming into the server is independent.

So we need a wrapper around pipes and sockets to separate out requests from substrate and this will give us a single model which is consistent for HTTP, sockets and pipes.

Protocol

On top of the medium is the protocol. This is the layer that deserialises bytes into the structured data that requests and responses consist of.

Request/Response processing

On top of the protocol is the logic for processing requests (on the server) or responses (on the client).

Server-side

Sketch:

MEDIUM  (factory for protocol, reads bytes & pushes to protocol,
         uses protocol to detect end-of-request, sends written
         bytes to client) e.g. socket, pipe, HTTP request handler.
 ^
 | bytes.
 v

PROTOCOL(serialization, deserialization)  accepts bytes for one
         request, decodes according to internal state, pushes
         structured data to handler.  accepts structured data from
         handler and encodes and writes to the medium.  factory for
         handler.
 ^
 | structured data
 v

HANDLER  (domain logic) accepts structured data, operates state
         machine until the request can be satisfied,
         sends structured data to the protocol.

Request handlers are registered in the bzrlib.smart.request module.

Client-side

Sketch:

CLIENT   domain logic, accepts domain requests, generated structured
         data, reads structured data from responses and turns into
         domain data.  Sends structured data to the protocol.
         Operates state machines until the request can be delivered
         (e.g. reading from a bundle generated in bzrlib to deliver a
         complete request).

         This is RemoteBzrDir, RemoteRepository, etc.
 ^
 | structured data
 v

PROTOCOL  (serialization, deserialization)  accepts structured data for one
         request, encodes and writes to the medium.  Reads bytes from the
         medium, decodes and allows the client to read structured data.
 ^
 | bytes.
 v

MEDIUM   accepts bytes from the protocol & delivers to the remote server.
         Allows the protocol to read bytes e.g. socket, pipe, HTTP request.

The domain logic is in bzrlib.remote: RemoteBzrDir, RemoteBranch, and so on.

There is also an plain file-level transport that calls remote methods to manipulate files on the server in bzrlib.transport.remote.

Protocol description

Version one

Version one of the protocol was introduced in Bazaar 0.11.

The protocol (for both requests and responses) is described by:

REQUEST := MESSAGE_V1
RESPONSE := MESSAGE_V1
MESSAGE_V1 := ARGS [BODY]

ARGS := ARG [MORE_ARGS] NEWLINE
MORE_ARGS := SEP ARG [MORE_ARGS]
SEP := 0x01

BODY := LENGTH NEWLINE BODY_BYTES TRAILER
LENGTH := decimal integer
TRAILER := "done" NEWLINE

That is, a tuple of arguments separated by Ctrl-A and terminated with a newline, followed by length prefixed body with a constant trailer. Note that although arguments are not 8-bit safe (they cannot include 0x01 or 0x0a bytes without breaking the protocol encoding), the body is.

Version two

Version two was introduced in Bazaar 0.16.

The request protocol is:

REQUEST_V2 := "bzr request 2" NEWLINE MESSAGE_V2

The response protocol is:

RESPONSE_V2 := "bzr response 2" NEWLINE RESPONSE_STATUS NEWLINE MESSAGE_V2
RESPONSE_STATUS := "success" | "failed"

Future versions should follow this structure, like version two does:

FUTURE_MESSAGE := VERSION_STRING NEWLINE REST_OF_MESSAGE

This is so that clients and servers can read bytes up to the first newline byte to determine what version a message is.

For compatibility will all versions (past and future) of bzr clients, servers that receive a request in an unknown protocol version should respond with a single-line error terminated with 0x0a (NEWLINE), rather than structured response prefixed with a version string.

Version two of the message protocol is:

MESSAGE_V2 := ARGS [BODY_V2]
BODY_V2 := BODY | STREAMED_BODY

That is, a version one length-prefixed body, or a version two streamed body.

Version two with streamed bodies

An extension to version two allows streamed bodies. A streamed body looks a lot like HTTP's chunked encoding:

STREAMED_BODY := "chunked" NEWLINE CHUNKS TERMINATOR
CHUNKS := CHUNK [CHUNKS]
CHUNK := HEX_LENGTH CHUNK_CONTENT
HEX_LENGTH := HEX_DIGITS NEWLINE
CHUNK_CONTENT := bytes

TERMINATOR := SUCCESS_TERMINATOR | ERROR_TERMINATOR
SUCCESS_TERMINATOR := 'END' NEWLINE
ERROR_TERMINATOR := 'ERR' NEWLINE CHUNKS SUCCESS_TERMINATOR

That is, the body consists of a series of chunks. Each chunk starts with a length prefix in hexadecimal digits, followed by an ASCII newline byte. The end of the body is signaled by 'END\\n', or by 'ERR\\n' followed by error args, one per chunk. Note that these args are 8-bit safe, unlike request args.

A streamed body starts with the string "chunked" so that legacy clients and servers will not mistake the first chunk as the start of a version one body.

The type of body (length-prefixed or chunked) in a response is always the same for a given request method. Only new request methods introduced in Bazaar 0.91 and later use streamed bodies.

Version three

Note

For some discussion of the requirements that led to this new protocol version, see bug #83935.

Version three has bencoding of most protocol structures, to make parsing simpler. For extra parsing convenience, these structures are length prefixed:

LENGTH_PREFIX := 32-bit unsigned integer in network byte order

Unlike earlier versions, clients and servers are no longer required to know which request verbs and responses will have bodies attached. Because of length-prefixing and other changes, it is always possible to know when a complete request or response has been read, even if the server implements no verbs.

The underlying message format is:

MESSAGE := MAGIC NEWLINE HEADERS CONTENTS END_MESSAGE
MAGIC := "bzr message 3 (bzr 1.6)"
HEADERS := LENGTH_PREFIX bencoded_dict
END_MESSAGE := "e"

BODY := MESSAGE_PART+
MESSAGE_PART := ONE_BYTE | STRUCTURE | BYTES
ONE_BYTE := "o" byte
STRUCTURE := "s" LENGTH_PREFIX bencoded_structure
BYTES := "b" LENGTH_PREFIX bytes

(Where + indicates one or more.)

This format allows an arbitrary sequence of message parts to be encoded in a single message. The contents of a MESSAGE have a higher-level message, but knowing just this amount of data it's possible to deserialize and consume a message, so that implementations can respond to messages sent by later versions.

Headers

Each request and response will have “headers”, a dictionary of key-value pairs. The keys must be strings, not any other type of value.

Currently, the only defined header is “Software version”. Both the client and the server should include a “Software version” header, with a value of a free-form string such as “bzrlib 1.5”, to aid debugging and logging. Clients and servers should not vary behaviour based on this string.

Conventional requests and responses

By convention, most requests and responses have a simple “arguments plus optional body” structure, as in earlier protocol versions. This section describes how such messages are encoded. All requests and responses defined by earlier protocol versions must be encoded in this way.

Conventional requests will send a CONTENTS of

CONV_REQ := ARGS SINGLE_OR_STREAMED_BODY?
SINGLE_OR_STREAMED_BODY := BYTES
      | BYTES+ TRAILER

ARGS := STRUCTURE(argument_tuple)
TRAILER := SUCCESS_STATUS | ERROR
SUCCESS_STATUS := ONE_BYTE("S")
ERROR := ONE_BYTE("E") STRUCTURE(argument_tuple)

Conventional responses will send CONTENTS of

CONV_RESP := RESP_STATUS ARGS SINGLE_OR_STREAMED_BODY?
RESP_STATUS := ONE_BYTE("S") | ONE_BYTE("E")

If the RESP_STATUS is success ("S"), the arguments are the method-dependent result.

For errors (where the Status byte of a response or a streamed body is "E"), the situation is analagous to requests. The first item in the encoded sequence must be a string of the error name. The other arguments supply details about the error, and their number and types will depend on the type of error (as identified by the error name).

Note that the streamed body from version two is now just multiple BYTES parts.

The end of the request or response is indicated by the lower-level END_MESSAGE. If there's only one BYTES element in the body, the TRAILER may or may not be present, depending on whether it was sent as a single chunk or as a stream that happens to have one element.

(Discussion) The success marker at the end of a streamed body seems redundant; it doesn't have space for any arguments, and the end of the body is marked anyhow by the end of the message. Recipients shouldn't take any action on it, though they should map an error into raising an error locally.

1.10 clients don't assert that they get a status byte at the end of the message. They will complain (in ConventionalResponseHandler.byte_part_received) if they get an initial success and then another byte part with no intervening bytes. If we stop sending the final success message and only flag errors they'll only get one if the error is detected after streaming starts but before any bytes are actually sent. Possibly we should wait until at least the first chunk is ready before declaring success.

For new methods, these sequences are just a convention and may be varied if appropriate for a particular request or response. However, each request should at least start with a STRUCTURE encoding the arguments tuple. The first element of that tuple must be a string that names the request method. (Note that arguments in this protocol version are bencoded. As a result, unlike previous protocol versions, arguments in this version are 8-bit clean.)

(Discussion) We're discussing having the byte segments be not just a method for sending a stream across the network, but actually having them be preserved in the rpc from end to end. This may be useful when there's an iterator on one side feeding in to an iterator on the other, if it avoids doing chunking and byte-counting at two levels, and if those iterators are a natural place to get good granularity. Also, for cases like insert_record_stream the server can't do much with the data until it gets a whole chunk, and so it'll be natural and efficient for it to be called with one chunk at a time.

On the other hand, there may be times when we've got some bytes from the network but not a full chunk, and it might be worthwhile to pass it up. If we promise to preserve chunks, then to do this we'd need two separate streaming interfaces: "we got a chunk" and "we got some bytes but not yet a full chunk". For insert_record_stream the second might not be useful, but it might be good when writing to a file where any number of bytes can be processed.

If we promise to preserve chunks, it'll tend to make some RPCs work only in chunks, and others just on whole blocks, and we can't so easily migrate RPCs from one to the other transparently to older implementations.

The data inside those chunks will be serialized anyhow, and possibly the data inside them will already be able to be serialized apart without understanding the chunks. Also, we might want to use these formats e.g. for pack files or in bundles, and so they don't particularly need lower-level chunking. So the current (unmerged, unstable) record stream serialization turns each record into a bencoded tuple and it'd be feasible to parse one tuple at a time from a byte stream that contains a sequence of them.

So we've decided that the chunks won't be semantic, and code should not count on them being preserved from client to server.

Early error returns

(Discussion) It would be nice if the server could notify the client of errors even before a streaming request has finished. This could cover situtaions such as the server not understanding the request, it being unable to open the requested location, or it finding that some of the revisions being sent are not actually needed.

Especially in the last case, we'd like to be able to gracefully notice the condition while the client is writing, and then have it adapt its behaviour. In any case, we don't want to have drop and restart the network stream.

It should be possible for the client to finish its current chunk and then its message, possibly with an error to cancel what's already been sent.

This relies on the client being able to read back from the server while it's writing. This is technically difficult for http but feasible over a socket or ssh.

We'd need a clean way to pass this back to the request method, even though it's presumably in the middle of doing its body iterator. Possibly the body iterator could be manually given a reference to the request object, and it can poll it to see if there's a response.

Perhaps we need to distinguish error conditions, which should turn into a client-side error regardless of the request code, from early success, which should be handled only if the request code specifically wants to do it.

Full-duplex operation

Code not geared to do pipelined requests, and this might require doing asynchrony within bzrlib. We might want to either go fully pipelined and asynchronous, but there might be a profitable middle ground.

The particular case where duplex communication would be good is in working towards the common points in the graphs between the client and server: we want to send speculatively, but detect as soon as they've matched up.

So we could for instance have a synchronous core, but rely on the OS network buffering to allow us to work on batches of say 64kB. We can also pipeline requests and responses, without allowing for them happening out of order, or mixed requests happening at the same time.

Wonder how our network performance would have turned out now if we'd done full-duplex from the start, and ignored hpss over http. We have pretty good (readonly) http support just over dumb http, and that may be better for many users.

APIs

On the client, the bzrlib code is "in charge": when it makes a request, or asks from data from the network, that causes network IO. The server is event driven: the network code tells the response handler when data has been received, and it takes back a Response object from the request handler that is then polled for body stream data.

Paths

Paths are passed across the network. The client needs to see a namespace that includes any repository that might need to be referenced, and the client needs to know about a root directory beyond which it cannot ascend.

Servers run over ssh will typically want to be able to access any path the user can access. Public servers on the other hand (which might be over http, ssh or tcp) will typically want to restrict access to only a particular directory and its children, so will want to do a software virtual root at that level. In other words they'll want to rewrite incoming paths to be under that level (and prevent escaping using ../ tricks). The default implementation in bzrlib does this using the bzrlib.transport.chroot module.

URLs that include ~ should probably be passed across to the server verbatim and the server can expand them. This will proably not be meaningful when limited to a directory? See bug 109143.

Requests

The first argument of a request specifies the request method.

The available request methods are registered in bzrlib.smart.request.

XXX: ideally the request methods should be documented here. Contributions welcome!

Recognised errors

The first argument of an error response specifies the error type.

One possible error name is UnknownMethod, which means the server does not recognise the verb used by the client's request. This error was introduced in version three.

XXX: ideally the error types should be documented here. Contributions welcome!