Solution: setsockopt returns EINVAL if the connection was closed by
the peer after the accept returned a valid socket. This is a valid
network error and should not cause an assert.
To handle this we have to extract the setsockopt from the stream
engine, as there's no clean way to return an error from the
constructor. Instead, try to set this option before creating the
engine in the callers, and return immediately as if the accept
had failed to avoid churn. Do the same for the connect calls by
setting the option in open_socket, so that the option for that
case is set even before connecting, so there's no possible race
condition.
Since this has to be done in 4 places (tcp/ipc listener, socks
connecter and open_socket) add an utility function in ip.cpp.
Fixes#1442
The mechanism is instanciated during the handshake itself, when and
error happen before this, the error method shall work anyway.
An error handling with a NULL mechanism means the handshake fail, so the
handshake failure event is also raised in this case.
- removed the previously added encryption_error, less changes less bug
- handshake fail is now signaled when an error happen while the
mechanism is still hanshaking
Solution: check if IPV6_TCLASS is defined so that when Hurd adds
support it will just work. Also it will avoid tripping over this on
other similar legacy systems.
Solution: wrap the event triggering in the DRAFT ifdef as well as the
defines. This ensures that the event are returned only if the
library was built with DRAFTs enabled.
Also update the test case to expect the new events since it uses
the catch-all mask. Although the sequence of event is different and
this might be considered as an API breakage, using the catch-all
ZMQ_EVENT_ALL mask implies that, well, all events are monitored so
it's normal that new ones will be added.
Users that don't want this kind of behaviour to change can simply
monitor only the event that they care about.
- Moved new events in draft section + added to zmq_draft.h
- Removed the remainning tabs
- Reverted the hard error (back to soft error) in curve_server.cpp
=> The feature doesn't works anymore
Added two new monitoring events:
- ZMQ_EVENT_HANDSHAKE_SUCCEED is raised once the encryption handshake succeed
- ZMQ_EVENT_HANDSHAKE_FAILED is raised when it failed
Both events are raised on server and client side.
* Fixed handling of WSAEWOULDBLOCK to be generic
I don't know what was the intention of this early if statement but
now this is properly evaluated in wsa_error_no function if this is
performance issue I suggest moving evaluating this error code to the
beginning of wsa_error_no.
* Fixed handling of WSAEWOULDBLOCK to be generic
Introduced default pointer to const char * and overrides this as NULL
if function is called by zmq::wsa_error()
* Fixed handling of WSAEWOULDBLOCK to be generic
Introduced default pointer to const char * and overrides this as NULL
if function is called by zmq::wsa_error()
Solution: fix the check for the socket.
This regression happens when using zloop with zmq_pollitem_it with
only file descriptors registerted through zloop_poller.
On native Windows, the file system is case-insensitive, so this this
doesn't matter there.
However, when compiling on a case-sensitive filesystem, such as when
cross-compiling from a Unixoid using mingw, we have to use the case
the file is supplied with, which is all lowercase.
Solution: restore inclusion of poll.h if using poll before zmq.h as
it was originally, as AIX redefines the POSIX structures and provides
compatibility macros.
Also add alternative aliases for 32 bit AIX's pollitem struct:
events -> reqevents
revents -> rtnevents
1. stl container is not thread safety
2. rm_fd() and loop() end to clear the retired event source will in multi-thread operation
3. may be crashed in reaper thread to delete the items in the std::vector as the source is nullptr
1. stl container is not thread safety
2. rm_fd() and loop() end to clear the retired event source will in multi-thread operation
3. may be crashed in reaper thread to delete the items in the std::vector as the source is nullptr
* bug fix: AIX pollset error
buf fix: AIX only, pollset 'rm_fd' set fd_table to null first then set pe->fd to retired_fd
* remove extra spaces
remove extra spaces
Summary:
To set `Type Of Service` for IP layer packets ZMQ provides `ZMQ_TOS` socket
option. However this only works for v4 sockets. Considering things are moving to
IPv6 heavily (especially within enterprise networks), ZMQ should support setting
`traffic class` for v6 based on `ZMQ_TOS`.
There is a subtle difference between v4 and v6 in terms of the positioning of
field but TOS has same meaning in both v4 and v6. Linux provides following APIs
for v4/v6 to set TOS field value.
```
// For v4
setsockopt(fd, IPPROTO_IP, IP_TOS, tos, sizeof(tos));
// For v6
setsockopt(fd, IPPROTO_IPV6, IPV6_TCLASS, tos, sizeof(tos));
```
Test Plan:
Make sure Cmake works fine and all tests passes.
Imported this change to `OpenR` project and tested on our local testbed.
Captured some packets exchanged between PUB/SUB and ROUTER/ROUTER sockets
pairs. Verify that `TCLASS` value is set as per expectation.
Tasks: #2208
#define ZMQ_HAVE_WINDOWS_TARGET_XP disable uncompatible WinAPI
1. Disable call if_indextoname()
2. Emulate windows Condition Variable API in class condition_variable_t with std::condition_variable
This code can be compiled in MSVC 2015 with option "Platform toolset: Visual Studio 2015 - Windows XP (v140_xp)"
Solution: Rename the custom implementation of clock_gettime for macOS to
alt_clock_gettime and wrap all usage in preprocessor macros to only enable the
alternative version when using macOS <= 10.11.
This issue occurs when targeting macOS 10.11 or earlier but using the 10.12
or newer SDK.
Solution: during a connect with a TCP endpoint if a source address is
passed set the SO_REUSEADDR flag on the socket before the bind system
call.
Add unit test to cover this case for both IPv4 and IPv6.
* - Fixed windows build errors
- Extended monitor lock scope to prevent race-condition between
process_stop and monitor
* - Fixed windows build errors
- Extended monitor lock scope to prevent race-condition between
process_stop and monitor
This should restore full compatibility with earlier zmq_poll behavior.
It complicates things a little bit, as collisions must be detected, and when collisions are found:
- event masks must be merged
- pollitems, events arrays are no longer co-ordered
Reverts the recent zmq_proxy patch to workaround the lack of repeat-item support in zmq_poll that is now fixed.
Return value is the number of events found. This also propagates to the return value of zmq_poller_wait_all.
zmq_poller_wait was only returning events on the first-registered socket.
Solution: use only Libs.private to avoid breaking application builds.
Even though Requires.private are supposed to be parsed only if
pkg-config is called with --static, the --cflags parameter is enough
to trigger the parsing, causing build failures for applications that
do not (and should not) depend on libzmq's dependencies.
Solution: add dependencies, if necessary, to the .private Libs and
Requires field of the pkgconfig file at build time.
This way pkg-config --static --libs libzmq will correctly print
dependencies if they were used to build the static libzmq.a library.
instead of allocating a new, identical array and copying the data.
This is only safe while zmq_poller_event_t and zmq::socket_poller_t::event_t are the same struct,
which they presumably will remain.
Solution: zmq_poller_wait_all signals all events
allows signaling multiple events with one call to zmq_poller_wait_all
rather than emitting only one event.
this prepares for zmq_poll being based on zmq_poller,
which requires events for all sockets rather than just one.
* fix bugs of the pollset
1. extend 'fd_table' when fd_ is greater or equal than the size of 'fd_table';
2. delete specific fd from pollset before reset pollin or pollout according the description of AIX document
* fix bugs of the pollset
edit error. remove extra spaces and paste fault
* fix bugs of pollset
remove character '-' at the end line.
Solution: add a zmq_assert to check if the ephemeral sockets created
to drain the queue of pending inproc connecting sockets was allocated
successfully.
Solution: check if the connecting inproc socket has been closed
before trying to send the identity.
Otherwise the pipe will be in waiting_for_delimiter state causing
writes to fail and the connect to assert when the context is being
torn down and the pending inproc connects are resolved.
Add test case that covers this behaviour.
Solution: allow for '[' character when doing the basic sanity check
on the TCP endpoint.
Also add unit tests for both IPv4 and IPv6 source;dest format.
Solution: Use only lower case for header file name.
We can find "wincrypt.h" by "WinCrypt.h" on Windows because Windows uses
case insensitive file system. But we can't find "wincrypt.h" by
"WinCrypt.h" on Linux Because Linux uses case sensitive file system.
The gssapi has some helper functions gssalloc_malloc()/gssalloc_free()
which on windows doesn't call malloc()/free(). Instead these are
wrappers around HeapAlloc() and HeapFree(). To complicate matters
gssapi doesn't export these helper functions, so you're left using
the allocation method of your choice.
See Here:
89683d1f13/src/lib/gssapi/generic/gssapi_alloc.h
The zmq gssapi implementation is calling malloc and then calling
gss_release_buffer() to free the memory. gss_release_buffer uses
gssalloc_free() to free this buffer which on windows calls HeapFree()
instead of free(). This causes an access violation on windows.
Linux provides accept4(2) which will return a socket with FD_CLOEXEC set
when called with the SOCK_CLOEXEC flag. So call this when available and
fall back to fcntl(..., FD_CLOEXEC) if not.
getifaddrs() can fail transiently with ECONNREFUSED on Linux.
This has been observed with Linux 3.10 when multiple processes
call zmq::tcp_address_t::resolve_nic_name() simultaneously.
Before asserting in this case, make 10 attempts, with exponential
backoff, given by (1 msec * 2^i), where i is the attempt number.
Fixes#2051
