4 There's a helper api `lws_daemonize` built by default that does everything you
5 need to daemonize well, including creating a lock file. If you're making
6 what's basically a daemon, just call this early in your init to fork to a
7 headless background process and exit the starting process.
9 Notice stdout, stderr, stdin are all redirected to /dev/null to enforce your
10 daemon is headless, so you'll need to sort out alternative logging, by, eg,
14 Maximum number of connections
15 -----------------------------
17 The maximum number of connections the library can deal with is decided when
18 it starts by querying the OS to find out how many file descriptors it is
19 allowed to open (1024 on Fedora for example). It then allocates arrays that
20 allow up to that many connections, minus whatever other file descriptors are
21 in use by the user code.
23 If you want to restrict that allocation, or increase it, you can use ulimit or
24 similar to change the avaiable number of file descriptors, and when restarted
25 **libwebsockets** will adapt accordingly.
28 Libwebsockets is singlethreaded
29 -------------------------------
31 Directly performing websocket actions from other threads is not allowed.
32 Aside from the internal data being inconsistent in `forked()` processes,
33 the scope of a `wsi` (`struct websocket`) can end at any time during service
34 with the socket closing and the `wsi` freed.
36 Websocket write activities should only take place in the
37 `LWS_CALLBACK_SERVER_WRITEABLE` callback as described below.
39 Only live connections appear in the user callbacks, so this removes any
40 possibility of trying to used closed and freed wsis.
42 If you need to service other socket or file descriptors as well as the
43 websocket ones, you can combine them together with the websocket ones
44 in one poll loop, see "External Polling Loop support" below, and
45 still do it all in one thread / process context.
47 If you insist on trying to use it from multiple threads, take special care if
48 you might simultaneously create more than one context from different threads.
50 SSL_library_init() is called from the context create api and it also is not
51 reentrant. So at least create the contexts sequentially.
54 Only send data when socket writeable
55 ------------------------------------
57 You should only send data on a websocket connection from the user callback
58 `LWS_CALLBACK_SERVER_WRITEABLE` (or `LWS_CALLBACK_CLIENT_WRITEABLE` for
61 If you want to send something, do not just send it but request a callback
62 when the socket is writeable using
64 - `lws_callback_on_writable(context, wsi)`` for a specific `wsi`, or
65 - `lws_callback_on_writable_all_protocol(protocol)` for all connections
66 using that protocol to get a callback when next writeable.
68 Usually you will get called back immediately next time around the service
69 loop, but if your peer is slow or temporarily inactive the callback will be
70 delayed accordingly. Generating what to write and sending it should be done
71 in the ...WRITEABLE callback.
73 See the test server code for an example of how to do this.
76 Do not rely on only your own WRITEABLE requests appearing
77 ---------------------------------------------------------
79 Libwebsockets may generate additional `LWS_CALLBACK_CLIENT_WRITEABLE` events
80 if it met network conditions where it had to buffer your send data internally.
82 So your code for `LWS_CALLBACK_CLIENT_WRITEABLE` needs to own the decision
83 about what to send, it can't assume that just because the writeable callback
84 came it really is time to send something.
86 It's quite possible you get an 'extra' writeable callback at any time and
87 just need to `return 0` and wait for the expected callback later.
90 Closing connections from the user side
91 --------------------------------------
93 When you want to close a connection, you do it by returning `-1` from a
94 callback for that connection.
96 You can provoke a callback by calling `lws_callback_on_writable` on
97 the wsi, then notice in the callback you want to close it and just return -1.
98 But usually, the decision to close is made in a callback already and returning
101 If the socket knows the connection is dead, because the peer closed or there
102 was an affirmitive network error like a FIN coming, then **libwebsockets** will
103 take care of closing the connection automatically.
105 If you have a silently dead connection, it's possible to enter a state where
106 the send pipe on the connection is choked but no ack will ever come, so the
107 dead connection will never become writeable. To cover that, you can use TCP
108 keepalives (see later in this document)
114 To support fragmented messages you need to check for the final
115 frame of a message with `lws_is_final_fragment`. This
116 check can be combined with `libwebsockets_remaining_packet_payload`
117 to gather the whole contents of a message, eg:
120 case LWS_CALLBACK_RECEIVE:
122 Client * const client = (Client *)user;
123 const size_t remaining = lws_remaining_packet_payload(wsi);
125 if (!remaining && lws_is_final_fragment(wsi)) {
126 if (client->HasFragments()) {
127 client->AppendMessageFragment(in, len, 0);
128 in = (void *)client->GetMessage();
129 len = client->GetMessageLength();
132 client->ProcessMessage((char *)in, len, wsi);
133 client->ResetMessage();
135 client->AppendMessageFragment(in, len, remaining);
140 The test app libwebsockets-test-fraggle sources also show how to
141 deal with fragmented messages.
147 Also using `lws_set_log_level` api you may provide a custom callback to actually
148 emit the log string. By default, this points to an internal emit function
149 that sends to stderr. Setting it to `NULL` leaves it as it is instead.
151 A helper function `lwsl_emit_syslog()` is exported from the library to simplify
152 logging to syslog. You still need to use `setlogmask`, `openlog` and `closelog`
155 The logging apis are made available for user code.
163 The difference between notice and info is that notice will be logged by default
164 whereas info is ignored by default.
167 External Polling Loop support
168 -----------------------------
170 **libwebsockets** maintains an internal `poll()` array for all of its
171 sockets, but you can instead integrate the sockets into an
172 external polling array. That's needed if **libwebsockets** will
173 cooperate with an existing poll array maintained by another
176 Four callbacks `LWS_CALLBACK_ADD_POLL_FD`, `LWS_CALLBACK_DEL_POLL_FD`,
177 `LWS_CALLBACK_SET_MODE_POLL_FD` and `LWS_CALLBACK_CLEAR_MODE_POLL_FD`
178 appear in the callback for protocol 0 and allow interface code to
179 manage socket descriptors in other poll loops.
181 You can pass all pollfds that need service to `lws_service_fd()`, even
182 if the socket or file does not belong to **libwebsockets** it is safe.
184 If **libwebsocket** handled it, it zeros the pollfd `revents` field before returning.
185 So you can let **libwebsockets** try and if `pollfd->revents` is nonzero on return,
186 you know it needs handling by your code.
189 Using with in c++ apps
190 ----------------------
192 The library is ready for use by C++ apps. You can get started quickly by
193 copying the test server
196 $ cp test-server/test-server.c test.cpp
199 and building it in C++ like this
202 $ g++ -DINSTALL_DATADIR=\"/usr/share\" -ocpptest test.cpp -lwebsockets
205 `INSTALL_DATADIR` is only needed because the test server uses it as shipped, if
206 you remove the references to it in your app you don't need to define it on
210 Availability of header information
211 ----------------------------------
213 From v1.2 of the library onwards, the HTTP header content is `free()`d as soon
214 as the websocket connection is established. For websocket servers, you can
215 copy interesting headers by handling `LWS_CALLBACK_FILTER_PROTOCOL_CONNECTION`
216 callback, for clients there's a new callback just for this purpose
217 `LWS_CALLBACK_CLIENT_FILTER_PRE_ESTABLISH`.
223 It is possible for a connection which is not being used to send to die
224 silently somewhere between the peer and the side not sending. In this case
225 by default TCP will just not report anything and you will never get any more
226 incoming data or sign the link is dead until you try to send.
228 To deal with getting a notification of that situation, you can choose to
229 enable TCP keepalives on all **libwebsockets** sockets, when you create the
232 To enable keepalive, set the ka_time member of the context creation parameter
233 struct to a nonzero value (in seconds) at context creation time. You should
234 also fill ka_probes and ka_interval in that case.
236 With keepalive enabled, the TCP layer will send control packets that should
237 stimulate a response from the peer without affecting link traffic. If the
238 response is not coming, the socket will announce an error at `poll()` forcing
241 Note that BSDs don't support keepalive time / probes / interval per-socket
242 like Linux does. On those systems you can enable keepalive by a nonzero
243 value in `ka_time`, but the systemwide kernel settings for the time / probes/
244 interval are used, regardless of what nonzero value is in `ka_time`.
246 Optimizing SSL connections
247 --------------------------
249 There's a member `ssl_cipher_list` in the `lws_context_creation_info` struct
250 which allows the user code to restrict the possible cipher selection at
251 context-creation time.
253 You might want to look into that to stop the ssl peers selecting a cipher which
254 is too computationally expensive. To use it, point it to a string like
256 `"RC4-MD5:RC4-SHA:AES128-SHA:AES256-SHA:HIGH:!DSS:!aNULL"`
258 if left `NULL`, then the "DEFAULT" set of ciphers are all possible to select.
261 Async nature of client connections
262 ----------------------------------
264 When you call `lws_client_connect(..)` and get a `wsi` back, it does not
265 mean your connection is active. It just mean it started trying to connect.
267 Your client connection is actually active only when you receive
268 `LWS_CALLBACK_CLIENT_ESTABLISHED` for it.
270 There's a 5 second timeout for the connection, and it may give up or die for
271 other reasons, if any of that happens you'll get a
272 `LWS_CALLBACK_CLIENT_CONNECTION_ERROR` callback on protocol 0 instead for the
275 After attempting the connection and getting back a non-`NULL` `wsi` you should
276 loop calling `lws_service()` until one of the above callbacks occurs.
278 As usual, see [test-client.c](test-server/test-client.c) for example code.