1ZMQ_SETSOCKOPT(3) 0MQ Manual ZMQ_SETSOCKOPT(3)
2
6 zmq_setsockopt - set 0MQ socket options
7
9 int zmq_setsockopt (void *socket, int option_name, const void
10 *option_value, size_t option_len);
11 Caution: All options, with the exception of ZMQ_SUBSCRIBE,
13 ZMQ_UNSUBSCRIBE and ZMQ_LINGER, only take effect for subsequent socket
14 bind/connects.
15
17 The zmq_setsockopt() function shall set the option specified by the
18 option_name argument to the value pointed to by the option_value
19 argument for the 0MQ socket pointed to by the socket argument. The
20 option_len argument is the size of the option value in bytes.
21 The following socket options can be set with the zmq_setsockopt()
23 function:
24 ZMQ_HWM: Set high water mark
26 The ZMQ_HWM option shall set the high water mark for the specified
27 socket. The high water mark is a hard limit on the maximum number of
28 outstanding messages 0MQ shall queue in memory for any single peer that
29 the specified socket is communicating with.
30 If this limit has been reached the socket shall enter an exceptional
32 state and depending on the socket type, 0MQ shall take appropriate
33 action such as blocking or dropping sent messages. Refer to the
34 individual socket descriptions in zmq_socket(3) for details on the
35 exact action taken for each socket type.
36 The default ZMQ_HWM value of zero means "no limit".
38 Option value type uint64_t
41 Option value unit messages
43 Default value 0
45 Applicable socket types all
47 ZMQ_SWAP: Set disk offload size
50 The ZMQ_SWAP option shall set the disk offload (swap) size for the
51 specified socket. A socket which has ZMQ_SWAP set to a non-zero value
52 may exceed it’s high water mark; in this case outstanding messages
53 shall be offloaded to storage on disk rather than held in memory.
54 The value of ZMQ_SWAP defines the maximum size of the swap space in
56 bytes.
57 Option value type int64_t
60 Option value unit bytes
62 Default value 0
64 Applicable socket types all
66 ZMQ_AFFINITY: Set I/O thread affinity
69 The ZMQ_AFFINITY option shall set the I/O thread affinity for newly
70 created connections on the specified socket.
71 Affinity determines which threads from the 0MQ I/O thread pool
73 associated with the socket’s context shall handle newly created
74 connections. A value of zero specifies no affinity, meaning that work
75 shall be distributed fairly among all 0MQ I/O threads in the thread
76 pool. For non-zero values, the lowest bit corresponds to thread 1,
77 second lowest bit to thread 2 and so on. For example, a value of 3
78 specifies that subsequent connections on socket shall be handled
79 exclusively by I/O threads 1 and 2.
80 See also zmq_init(3) for details on allocating the number of I/O
82 threads for a specific context.
83 Option value type uint64_t
86 Option value unit N/A (bitmap)
88 Default value 0
90 Applicable socket types N/A
92 ZMQ_IDENTITY: Set socket identity
95 The ZMQ_IDENTITY option shall set the identity of the specified socket.
96 Socket identity determines if existing 0MQ infrastructure (message
97 queues, forwarding devices) shall be identified with a specific
98 application and persist across multiple runs of the application.
99 If the socket has no identity, each run of an application is completely
101 separate from other runs. However, with identity set the socket shall
102 re-use any existing 0MQ infrastructure configured by the previous
103 run(s). Thus the application may receive messages that were sent in the
104 meantime, message queue limits shall be shared with previous run(s) and
105 so on.
106 Identity should be at least one byte and at most 255 bytes long.
108 Identities starting with binary zero are reserved for use by 0MQ
109 infrastructure.
110 Option value type binary data
113 Option value unit N/A
115 Default value NULL
117 Applicable socket types all
119 ZMQ_SUBSCRIBE: Establish message filter
122 The ZMQ_SUBSCRIBE option shall establish a new message filter on a
123 ZMQ_SUB socket. Newly created ZMQ_SUB sockets shall filter out all
124 incoming messages, therefore you should call this option to establish
125 an initial message filter.
126 An empty option_value of length zero shall subscribe to all incoming
128 messages. A non-empty option_value shall subscribe to all messages
129 beginning with the specified prefix. Multiple filters may be attached
130 to a single ZMQ_SUB socket, in which case a message shall be accepted
131 if it matches at least one filter.
132 Option value type binary data
135 Option value unit N/A
137 Default value N/A
139 Applicable socket types ZMQ_SUB
141 ZMQ_UNSUBSCRIBE: Remove message filter
144 The ZMQ_UNSUBSCRIBE option shall remove an existing message filter on a
145 ZMQ_SUB socket. The filter specified must match an existing filter
146 previously established with the ZMQ_SUBSCRIBE option. If the socket has
147 several instances of the same filter attached the ZMQ_UNSUBSCRIBE
148 option shall remove only one instance, leaving the rest in place and
149 functional.
150 Option value type binary data
153 Option value unit N/A
155 Default value N/A
157 Applicable socket types ZMQ_SUB
159 ZMQ_RATE: Set multicast data rate
162 The ZMQ_RATE option shall set the maximum send or receive data rate for
163 multicast transports such as zmq_pgm(7) using the specified socket.
164 Option value type int64_t
167 Option value unit kilobits per second
169 Default value 100
171 Applicable socket types all, when using multicast
173 transports
174 ZMQ_RECOVERY_IVL: Set multicast recovery interval
177 The ZMQ_RECOVERY_IVL option shall set the recovery interval for
178 multicast transports using the specified socket. The recovery interval
179 determines the maximum time in seconds that a receiver can be absent
180 from a multicast group before unrecoverable data loss will occur.
181 Caution
183 Exercise care when setting large recovery intervals as the data
184 needed for recovery will be held in memory. For example, a 1 minute
185 recovery interval at a data rate of 1Gbps requires a 7GB in-memory
186 buffer.
187 Option value type int64_t
190 Option value unit seconds
192 Default value 10
194 Applicable socket types all, when using multicast
196 transports
197 ZMQ_RECOVERY_IVL_MSEC: Set multicast recovery interval in milliseconds
200 The ZMQ_RECOVERY_IVL_MSEC option shall set the recovery interval,
201 specified in milliseconds (ms) for multicast transports using the
202 specified socket. The recovery interval determines the maximum time in
203 milliseconds that a receiver can be absent from a multicast group
204 before unrecoverable data loss will occur.
205 A non-zero value of the ZMQ_RECOVERY_IVL_MSEC option will take
207 precedence over the ZMQ_RECOVERY_IVL option, but since the default for
208 the ZMQ_RECOVERY_IVL_MSEC is -1, the default is to use the
209 ZMQ_RECOVERY_IVL option value.
210 Caution
212 Exercise care when setting large recovery intervals as the data
213 needed for recovery will be held in memory. For example, a 1 minute
214 recovery interval at a data rate of 1Gbps requires a 7GB in-memory
215 buffer.
216 Option value type int64_t
219 Option value unit milliseconds
221 Default value -1
223 Applicable socket types all, when using multicast
225 transports
226 ZMQ_MCAST_LOOP: Control multicast loop-back
229 The ZMQ_MCAST_LOOP option shall control whether data sent via multicast
230 transports using the specified socket can also be received by the
231 sending host via loop-back. A value of zero disables the loop-back
232 functionality, while the default value of 1 enables the loop-back
233 functionality. Leaving multicast loop-back enabled when it is not
234 required can have a negative impact on performance. Where possible,
235 disable ZMQ_MCAST_LOOP in production environments.
236 Option value type int64_t
239 Option value unit boolean
241 Default value 1
243 Applicable socket types all, when using multicast
245 transports
246 ZMQ_SNDBUF: Set kernel transmit buffer size
249 The ZMQ_SNDBUF option shall set the underlying kernel transmit buffer
250 size for the socket to the specified size in bytes. A value of zero
251 means leave the OS default unchanged. For details please refer to your
252 operating system documentation for the SO_SNDBUF socket option.
253 Option value type uint64_t
256 Option value unit bytes
258 Default value 0
260 Applicable socket types all
262 ZMQ_RCVBUF: Set kernel receive buffer size
265 The ZMQ_RCVBUF option shall set the underlying kernel receive buffer
266 size for the socket to the specified size in bytes. A value of zero
267 means leave the OS default unchanged. For details refer to your
268 operating system documentation for the SO_RCVBUF socket option.
269 Option value type uint64_t
272 Option value unit bytes
274 Default value 0
276 Applicable socket types all
278 ZMQ_LINGER: Set linger period for socket shutdown
281 The ZMQ_LINGER option shall set the linger period for the specified
282 socket. The linger period determines how long pending messages which
283 have yet to be sent to a peer shall linger in memory after a socket is
284 closed with zmq_close(3), and further affects the termination of the
285 socket’s context with zmq_term(3). The following outlines the different
286 behaviours:
287 · The default value of -1 specifies an infinite linger period.
289 Pending messages shall not be discarded after a call to
290 zmq_close(); attempting to terminate the socket’s context with
291 zmq_term() shall block until all pending messages have been sent to
292 a peer.
293 · The value of 0 specifies no linger period. Pending messages shall
295 be discarded immediately when the socket is closed with
296 zmq_close().
297 · Positive values specify an upper bound for the linger period in
299 milliseconds. Pending messages shall not be discarded after a call
300 to zmq_close(); attempting to terminate the socket’s context with
301 zmq_term() shall block until either all pending messages have been
302 sent to a peer, or the linger period expires, after which any
303 pending messages shall be discarded.
304 Option value type int
306 Option value unit milliseconds
307 Default value -1 (infinite)
308 Applicable socket types all
309 ZMQ_RECONNECT_IVL: Set reconnection interval
312 The ZMQ_RECONNECT_IVL option shall set the initial reconnection
313 interval for the specified socket. The reconnection interval is the
314 period 0MQ shall wait between attempts to reconnect disconnected peers
315 when using connection-oriented transports.
316 Note
318 The reconnection interval may be randomized by 0MQ to prevent
319 reconnection storms in topologies with a large number of peers per
320 socket.
321 Option value type int
324 Option value unit milliseconds
326 Default value 100
328 Applicable socket types all, only for
330 connection-oriented
331 transports
332 ZMQ_RECONNECT_IVL_MAX: Set maximum reconnection interval
335 The ZMQ_RECONNECT_IVL_MAX option shall set the maximum reconnection
336 interval for the specified socket. This is the maximum period 0MQ shall
337 wait between attempts to reconnect. On each reconnect attempt, the
338 previous interval shall be doubled untill ZMQ_RECONNECT_IVL_MAX is
339 reached. This allows for exponential backoff strategy. Default value
340 means no exponential backoff is performed and reconnect interval
341 calculations are only based on ZMQ_RECONNECT_IVL.
342 Note
344 Values less than ZMQ_RECONNECT_IVL will be ignored.
345 Option value type int
348 Option value unit milliseconds
350 Default value 0 (only use
352 ZMQ_RECONNECT_IVL)
353 Applicable socket types all, only for
355 connection-oriented
356 transports
357 ZMQ_BACKLOG: Set maximum length of the queue of outstanding connections
360 The ZMQ_BACKLOG option shall set the maximum length of the queue of
361 outstanding peer connections for the specified socket; this only
362 applies to connection-oriented transports. For details refer to your
363 operating system documentation for the listen function.
364 Option value type int
367 Option value unit connections
369 Default value 100
371 Applicable socket types all, only for
373 connection-oriented
374 transports.
375
378 The zmq_setsockopt() function shall return zero if successful.
379 Otherwise it shall return -1 and set errno to one of the values defined
380 below.
381
383 EINVAL
384 The requested option option_name is unknown, or the requested
385 option_len or option_value is invalid.
386 ETERM
388 The 0MQ context associated with the specified socket was
389 terminated.
390 EFAULT
392 The provided socket was not valid (NULL).
393 EINTR
395 The operation was interrupted by delivery of a signal.
396
398 Subscribing to messages on a ZMQ_SUB socket.
399 /* Subscribe to all messages */
401 rc = zmq_setsockopt (socket, ZMQ_SUBSCRIBE, "", 0);
402 assert (rc == 0);
403 /* Subscribe to messages prefixed with "ANIMALS.CATS" */
404 rc = zmq_setsockopt (socket, ZMQ_SUBSCRIBE, "ANIMALS.CATS", 12);
405 Setting I/O thread affinity.
407 int64_t affinity;
409 /* Incoming connections on TCP port 5555 shall be handled by I/O thread 1 */
410 affinity = 1;
411 rc = zmq_setsockopt (socket, ZMQ_AFFINITY, &affinity, sizeof affinity);
412 assert (rc);
413 rc = zmq_bind (socket, "tcp://lo:5555");
414 assert (rc);
415 /* Incoming connections on TCP port 5556 shall be handled by I/O thread 2 */
416 affinity = 2;
417 rc = zmq_setsockopt (socket, ZMQ_AFFINITY, &affinity, sizeof affinity);
418 assert (rc);
419 rc = zmq_bind (socket, "tcp://lo:5556");
420 assert (rc);
421
424 zmq_getsockopt(3) zmq_socket(3) zmq(7)
425
427 This 0MQ manual page was written by Martin Sustrik
428 <sustrik@250bpm.com[1]> and Martin Lucina <mato@kotelna.sk[2]>.
429
431 1. sustrik@250bpm.com
432 mailto:sustrik@250bpm.com
433 2. mato@kotelna.sk
435 mailto:mato@kotelna.sk
4360MQ 2.1.4 03/30/2011 ZMQ_SETSOCKOPT(3)