- 12 Jun, 2012 2 commits
-
-
Ian Barber authored
This patch, salvaged from a trainwreck accidental merge earlier, adds a new sockopt, ZMQ_DELAY_ATTACH_ON_CONNECT which prevents a end point being available to push messages to until it has fully connected, making connect work more like bind. This also applies to reconnecting sockets, which may cause message loss of in-queue messages, so it is sensible to use this in conjunction with a low HWM and potentially an alternative acknowledgement path. Notes on most of the individual commits can be found the repository log.
-
Ian Barber authored
Revert "After speaking with Ben Gray and the discussion on the mailing list, this is an attempt to create a sockopt to allow connecting pipes to not immediately be available for traffic. The problem is in a PUSH to many PULL situation, where there is a connect to a PULL which is not there. This connect will immediately create a pipe (unlike bind), and traffic will be load balanced to that pipe. This means if there is a persistently unavailable end point then the traffic will queue until HWM is hit, and older messages will be lost." This reverts commit fe3fb419.
-
- 06 Jun, 2012 1 commit
-
-
Pieter Hintjens authored
-
- 01 Jun, 2012 1 commit
-
-
Ian Barber authored
After speaking with Ben Gray and the discussion on the mailing list, this is an attempt to create a sockopt to allow connecting pipes to not immediately be available for traffic. The problem is in a PUSH to many PULL situation, where there is a connect to a PULL which is not there. This connect will immediately create a pipe (unlike bind), and traffic will be load balanced to that pipe. This means if there is a persistently unavailable end point then the traffic will queue until HWM is hit, and older messages will be lost. This patch adds a sockopt ZMQ_DELAY_ATTACH_ON_CONNECT, which if set to 1 will attempt to preempt this behavior. It does this by extending the use of the session_base to include in the outbound as well as the inbound pipe, and only associates the pipe with the socket once it receives the connected callback via a process_attach message. This works, and a test has been added to show so, but may introduce unexpected complications. The shutdown logic in this class has become marginally more awkward because of this, requiring the session to serve as the sink for both pipes if shutdown occurs with a still-connecting pipe in place. It is also possible there could be issues around flushing the messages, but as I could not directly think how to create such an issue I have not written any code with regards to that. The documentation has been updated to reflect the change, but please do check over the code and test and review.
-
- 22 May, 2012 1 commit
-
-
Lourens Naudé authored
-
- 21 May, 2012 1 commit
-
-
Lourens Naudé authored
-
- 20 May, 2012 2 commits
-
-
Lourens Naudé authored
Rename type zmq_monitor_fn -> zmq_monitor for a more natural callback definition API (zmq_monitor type, monitor.function callback)
-
Lourens Naudé authored
Change zmq_monitor_fn type to cast between pointer-to-object and pointer-to-function in a more standards compliant way
-
- 04 May, 2012 1 commit
-
-
Lourens Naudé authored
-
- 20 Apr, 2012 1 commit
-
-
Sergey KHripchenko authored
-
- 18 Apr, 2012 1 commit
-
-
Sergey KHripchenko authored
-
- 12 Apr, 2012 1 commit
-
-
Sergey KHripchenko authored
Assign arbitrary number of filters that will be applied for each new TCP transport connection on a listening socket. If no filters applied, then TCP transport allows connections from any ip. If at least one filter is applied then new connection source ip should be matched. To clear all filters call zmq_setsockopt(socket, ZMQ_TCP_ACCEPT_FILTER, NULL, 0). Filter is a null-terminated string with ipv6 or ipv4 CIDR. For example: localhost 127.0.0.1 mail.ru/24 ::1 ::1/128 3ffe:1:: 3ffe:1::/56 Returns -1 if the filter couldn't be assigned(format error or ipv6 filter with ZMQ_IPV4ONLY set) P.S. The only thing that worries me is that I had to re-enable 'default assign by reference constructor/operator' for 'tcp_address_t' (and for my inherited class tcp_address_mask_t) to store it in std::vector in 'options_t'...
-
- 10 Apr, 2012 1 commit
-
-
jdc8 authored
-
- 05 Apr, 2012 1 commit
-
-
Sergey KHripchenko authored
currently not fully cross-platform work on linux possibly work in *bsd and could be enhanced to work on windows
-
- 04 Apr, 2012 1 commit
-
-
Pieter Hintjens authored
-
- 21 Mar, 2012 1 commit
-
-
Pieter Hintjens authored
* Return integer property * Fixed up man pages, which were inaccurate * Fixed test case
-
- 20 Mar, 2012 1 commit
-
-
Pieter Hintjens authored
* Implemented new ctx API (_new, _destroy, _get, _set) * Removed 'typesafe' macros from zmq.h * Added support for MAX_SOCKETS (was tied into change for #337) * Created new man pages
-
- 16 Mar, 2012 1 commit
-
-
Pieter Hintjens authored
- reverted commit 941be8d2. - fixed zmq_device implementation for latest socket_base class - added back zmq_device.3 man page
-
- 15 Mar, 2012 2 commits
- 14 Mar, 2012 1 commit
-
-
boris@boressoft.ru authored
* Added two new files: errno.hpp and errno.cpp. They are required to use errno functionality on WM. * zmq.cpp, msg.h: removed inclusion of errno.h because it is included in zmq.h that is also included by .cpp. * windows.hpp: process.h is included only for desktop builds. * thread.cpp: on CE CreateThread is used instead of __beginthreadex * socket_base.cpp, clock.cpp: on CE include cmnintrin.h instead on intrin.h * signaler.cpp: on Windows should use special macro around event name (for unicode builds) * err.hpp: make it include errno.hpp (my file) instead on errno.h when building for CE * err.cpp: use FormatMessage when building for CE (because CE does not have ANSI API functions) * zmq.h: do not include errno.h whe building for CE * libzmq.vcproj: add tro new files
-
- 17 Feb, 2012 2 commits
-
-
skaller authored
-
Mikko Koppanen authored
-
- 16 Feb, 2012 6 commits
-
-
Pieter Hintjens authored
-
Pieter Hintjens authored
-
Pieter Hintjens authored
-
Pieter Hintjens authored
-
Pieter Hintjens authored
-
Pieter Hintjens authored
* Added zmq_msg_poke for orthogonality * Added zmq_msg_more for simplicity * Fixed up man pages and test program
-
- 15 Feb, 2012 4 commits
-
-
Pieter Hintjens authored
-
Pieter Hintjens authored
-
Pieter Hintjens authored
-
Pieter Hintjens authored
-
- 14 Feb, 2012 1 commit
-
-
Ian Barber authored
-
- 11 Feb, 2012 1 commit
-
-
skaller authored
-
- 08 Feb, 2012 3 commits
-
-
Ian Barber authored
-
skaller authored
We have to use an incomplete type in the interface. The definition of iovec is only in the implementation. This appears to following existing practice in 0MQ. It seems a bit flakey that zmq.h is not included in zmq.cpp, which is normal practice to ensure the implementation matches the specified interface. YMMV. I follow 0MQ style.
-
skaller authored
Also move iovec definition from implementation to interface. Not clear to me at present if an incomplete type should be used to avoid gratuitously including <sys/uio.h> in the interface. The interface can't be used with this include.
-
- 03 Feb, 2012 1 commit
-
-
skaller authored
We use a distinct context initialisation function to specify all sockets derived therefrom will be thread safe. However the inheritance is done exclusively in the C interface. This is not really correct, but it is chosen to minimise interference with the existing C++ code, including any construct or other calls within the C++ code base. Semantically the C++ code should be unchanged, physically some data structures and extra methods are provided by they're only used from the C binding.
-
- 19 Dec, 2011 1 commit
-
-
Martin Sustrik authored
Signed-off-by:
Martin Sustrik <sustrik@250bpm.com>
-