path: root/doc/xs_socket.txt

xs_socket(3)
============


NAME
----
xs_socket - create Crossroads socket


SYNOPSIS
--------
*void *xs_socket (void '*context', int 'type');*


DESCRIPTION
-----------
The 'xs_socket()' function shall create a Crossroads socket within the specified
'context' and return an opaque handle to the newly created socket. The 'type'
argument specifies the socket type, which determines the semantics of
communication over the socket.

The newly created socket is initially unbound, and not associated with any
endpoints. In order to establish a message flow a socket must first be
connected to at least one endpoint with linkxs:xs_connect[3], or at least one
endpoint must be created for accepting incoming connections with
linkxs:xs_bind[3].

.Key differences to conventional sockets
Generally speaking, conventional sockets present a _synchronous_ interface to
either connection-oriented reliable byte streams (SOCK_STREAM), or
connection-less unreliable datagrams (SOCK_DGRAM). In comparison, Crossroads
sockets present an abstraction of an asynchronous _message queue_, with the
exact queueing semantics depending on the socket type in use. Where conventional
sockets transfer streams of bytes or discrete datagrams, Crossroads sockets
transfer discrete _messages_.

Crossroads sockets being _asynchronous_ means that the timings of the physical
connection setup and tear down, reconnect and effective delivery are transparent
to the user and organized by Crossroads library itself. Further, messages may
be _queued_ in the event that a peer is unavailable to receive them.

Conventional sockets allow only strict one-to-one (two peers), many-to-one
(many clients, one server), or in some cases one-to-many (multicast)
relationships. With the exception of 'XS_PAIR', Crossroads sockets may be
connected *to multiple endpoints* using _xs_connect()_, while simultaneously
accepting incoming connections *from multiple endpoints* bound to the socket
using _xs_bind()_, thus allowing many-to-many relationships.

.Thread safety
Crossroads 'sockets' are _not_ thread safe. Applications MUST NOT use a socket
from multiple threads except after migrating a socket from one thread to 
another with a "full fence" memory barrier.

.Socket types
Crossroads defines several messaging patterns which encapsulate exact semantics
of a particular topology. For example, publish-subscribe pattern defines data
distribution trees while request-reply defines networks of shared stateless
services. Each pattern defines several socket types (roles in the pattern).

The following sections present the socket types defined by Crossroads library:


Request-reply pattern
~~~~~~~~~~~~~~~~~~~~~
The request-reply pattern is used for sending requests from a _client_ to one
or more instances of a stateless _service_, and receiving subsequent replies
to each request sent.


XS_REQ
^^^^^^
A socket of type 'XS_REQ' is used by a _client_ to send requests to and
receive replies from a _service_. This socket type allows only an alternating
sequence of _xs_send(request)_ and subsequent _xs_recv(reply)_ calls. Each
request sent is load-balanced among all _services_, and each reply received is
matched with the last issued request.

When a 'XS_REQ' socket enters an exceptional state due to having reached the
high water mark for all _services_, or if there are no _services_ at all, then
any linkxs:xs_send[3] operations on the socket shall block until the
exceptional state ends or at least one _service_ becomes available for sending;
messages are not discarded.

[horizontal]
.Summary of XS_REQ characteristics
Compatible peer sockets:: 'XS_REP'
Send/receive pattern:: Send, Receive, Send, Receive, ...
Outgoing routing strategy:: Load-balanced
Incoming routing strategy:: Last peer
XS_HWM option action:: Block


XS_REP
^^^^^^
A socket of type 'XS_REP' is used by a _service_ to receive requests from and
send replies to a _client_. This socket type allows only an alternating
sequence of _xs_recv(request)_ and subsequent _xs_send(reply)_ calls. Each
request received is fair-queued from among all _clients_, and each reply sent
is routed to the _client_ that issued the last request. If the original
requester doesn't exist any more the reply is silently discarded.

When a 'XS_REP' socket enters an exceptional state due to having reached the
high water mark for a _client_, then any replies sent to the _client_ in
question shall be dropped until the exceptional state ends.

[horizontal]
.Summary of XS_REP characteristics
Compatible peer sockets:: 'XS_REQ'
Send/receive pattern:: Receive, Send, Receive, Send, ...
Incoming routing strategy:: Fair-queued
Outgoing routing strategy:: Last peer
XS_HWM option action:: Drop


XS_XREQ
^^^^^^^
A socket of type 'XS_XREQ' is a socket type underlying 'XS_REQ'. It doesn't
impose the strict order of sends and recvs as 'XS_REQ' does and it is
intended for use in intermediate devices in request-reply topologies.

Each message sent is load-balanced among all connected
peers, and each message received is fair-queued from all connected peers.

When a 'XS_XREQ' socket enters an exceptional state due to having reached the
high water mark for all peers, or if there are no peers at all, then any
linkxs:xs_send[3] operations on the socket shall block until the exceptional
state ends or at least one peer becomes available for sending; messages are not
discarded.

[horizontal]
.Summary of XS_XREQ characteristics
Compatible peer sockets:: 'XS_XREP', 'XS_REP'
Send/receive pattern:: Unrestricted
Outgoing routing strategy:: Load-balanced
Incoming routing strategy:: Fair-queued
XS_HWM option action:: Block


XS_XREP
^^^^^^^
A socket of type 'XS_XREP' is a socket type underlying 'XS_REP'. It doesn't
impose the strict order of sends and recvs as 'XS_REQ' does and it is
intended for use in intermediate devices in request-reply topologies.

Messages received are fair-queued from among all connected peers. The outbound
messages are routed to a specific peer, as explained below.

When a 'XS_XREP' socket enters an exceptional state due to having reached the
high water mark for all peers, or if there are no peers at all, then any
messages sent to the socket shall be dropped until the exceptional state ends.
Likewise, any messages to be routed to a non-existent peer or a peer for which
the individual high water mark has been reached shall also be dropped.

[horizontal]
.Summary of XS_XREP characteristics
Compatible peer sockets:: 'XS_XREQ', 'XS_REQ'
Send/receive pattern:: Unrestricted
Outgoing routing strategy:: See text
Incoming routing strategy:: Fair-queued
XS_HWM option action:: Drop


Publish-subscribe pattern
~~~~~~~~~~~~~~~~~~~~~~~~~
The publish-subscribe pattern is used for one-to-many distribution of data from
a single _publisher_ to multiple _subscribers_ in a fan out fashion.


XS_PUB
^^^^^^
A socket of type 'XS_PUB' is used by a _publisher_ to distribute data.
Messages sent are distributed in a fan out fashion to all connected peers.
The linkxs:xs_recv[3] function is not implemented for this socket type.

When a 'XS_PUB' socket enters an exceptional state due to having reached the
high water mark for a _subscriber_, then any messages that would be sent to the
_subscriber_ in question shall instead be dropped until the exceptional state
ends. The _xs_send()_ function shall never block for this socket type.

[horizontal]
.Summary of XS_PUB characteristics
Compatible peer sockets:: 'XS_SUB', 'XS_XSUB'
Send/receive pattern:: Send only
Incoming routing strategy:: N/A
Outgoing routing strategy:: Fan out
XS_HWM option action:: Drop


XS_SUB
^^^^^^
A socket of type 'XS_SUB' is used by a _subscriber_ to subscribe to data
distributed by a _publisher_. Initially a 'XS_SUB' socket is not subscribed to
any messages, use the 'XS_SUBSCRIBE' option of linkxs:xs_setsockopt[3] to
specify which messages to subscribe to. The _xs_send()_ function is not
implemented for this socket type.

[horizontal]
.Summary of XS_SUB characteristics
Compatible peer sockets:: 'XS_PUB', 'XS_XPUB'
Send/receive pattern:: Receive only
Incoming routing strategy:: Fair-queued
Outgoing routing strategy:: N/A
XS_HWM option action:: Drop


XS_XPUB
^^^^^^^
Same as XS_PUB except that you can receive subscriptions from the peers
in form of incoming messages. Subscription message is a byte 1 (for
subscriptions) or byte 0 (for unsubscriptions) followed by the subscription
body.

[horizontal]
.Summary of XS_XPUB characteristics
Compatible peer sockets:: 'XS_SUB', 'XS_XSUB'
Send/receive pattern:: Send messages, receive subscriptions
Incoming routing strategy:: N/A
Outgoing routing strategy:: Fan out
XS_HWM option action:: Drop


XS_XSUB
^^^^^^^
Same as XS_SUB except that you subscribe by sending subscription messages to
the socket. Subscription message is a byte 1 (for subscriptions) or byte 0
(for unsubscriptions) followed by the subscription body.

[horizontal]
.Summary of XS_XSUB characteristics
Compatible peer sockets:: 'XS_PUB', 'XS_XPUB'
Send/receive pattern:: Receive messages, send subscriptions
Incoming routing strategy:: Fair-queued
Outgoing routing strategy:: N/A
XS_HWM option action:: Drop


Pipeline pattern
~~~~~~~~~~~~~~~~
The pipeline pattern is used for distributing data to _nodes_ arranged in
a pipeline. Data always flows down the pipeline, and each stage of the pipeline
is connected to at least one _node_. When a pipeline stage is connected to
multiple _nodes_ data is load-balanced among all connected _nodes_.


XS_PUSH
^^^^^^^
A socket of type 'XS_PUSH' is used by a pipeline _node_ to send messages
to downstream pipeline _nodes_. Messages are load-balanced to all connected
downstream _nodes_. The _xs_recv()_ function is not implemented for this
socket type.

When a 'XS_PUSH' socket enters an exceptional state due to having reached the
high water mark for all downstream _nodes_, or if there are no downstream
_nodes_ at all, then any linkxs:xs_send[3] operations on the socket shall
block until the exceptional state ends or at least one downstream _node_
becomes available for sending; messages are not discarded.

[horizontal]
.Summary of XS_PUSH characteristics
Compatible peer sockets:: 'XS_PULL'
Direction:: Unidirectional
Send/receive pattern:: Send only
Incoming routing strategy:: N/A
Outgoing routing strategy:: Load-balanced
XS_HWM option action:: Block


XS_PULL
^^^^^^^
A socket of type 'XS_PULL' is used by a pipeline _node_ to receive messages
from upstream pipeline _nodes_. Messages are fair-queued from among all
connected upstream _nodes_. The _xs_send()_ function is not implemented for
this socket type.

[horizontal]
.Summary of XS_PULL characteristics
Compatible peer sockets:: 'XS_PUSH'
Direction:: Unidirectional
Send/receive pattern:: Receive only
Incoming routing strategy:: Fair-queued
Outgoing routing strategy:: N/A
XS_HWM option action:: N/A


Exclusive pair pattern
~~~~~~~~~~~~~~~~~~~~~~
The exclusive pair is an advanced pattern used for communicating exclusively
between two peers.


XS_PAIR
^^^^^^^
A socket of type 'XS_PAIR' can only be connected to a single peer at any one
time.  No message routing or filtering is performed on messages sent over a
'XS_PAIR' socket.

When a 'XS_PAIR' socket enters an exceptional state due to having reached the
high water mark for the connected peer, or if no peer is connected, then
any linkxs:xs_send[3] operations on the socket shall block until the peer
becomes available for sending; messages are not discarded.

NOTE: 'XS_PAIR' sockets are experimental, and are currently missing several
features such as auto-reconnection.

[horizontal]
.Summary of XS_PAIR characteristics
Compatible peer sockets:: 'XS_PAIR'
Direction:: Bidirectional
Send/receive pattern:: Unrestricted
Incoming routing strategy:: N/A
Outgoing routing strategy:: N/A
XS_HWM option action:: Block


RETURN VALUE
------------
The _xs_socket()_ function shall return an opaque handle to the newly created
socket if successful. Otherwise, it shall return NULL and set 'errno' to one of
the values defined below.


ERRORS
------
*EINVAL*::
The requested socket 'type' is invalid.
*EFAULT*::
The provided 'context' is invalid.
*EMFILE*::
The limit on the total number of open Crossroads sockets has been reached.
*ETERM*::
The context specified was terminated.

SEE ALSO
--------
linkxs:xs_init[3]
linkxs:xs_setsockopt[3]
linkxs:xs_bind[3]
linkxs:xs_connect[3]
linkxs:xs_send[3]
linkxs:xs_recv[3]
linkxs:xs[7]


AUTHORS
-------
The Crossroads documentation was written by Martin Sustrik <sustrik@250bpm.com>
and Martin Lucina <martin@lucina.net>.