Pieter Hintjens
076e081de2
Merge pull request #373 from ianbarber/master
...
Add ZMQ_DELAY_ATTACH_ON_CONNECT sockopt
2012-06-12 14:18:51 -07:00
Pieter Hintjens
e1cc2d4a41
Merge pull request #372 from ianbarber/manfix
...
Update bind and connect documentation for clarity
2012-06-12 12:26:20 -07:00
Ian Barber
bc9ae715c3
Add asserts and rename pipe set
...
Rename the pipeset to terminating_pipes, as suggested by Martin H. Adds
asserts to test the pipe is contained in the terminating set where
appropriate.
2012-06-12 17:56:39 +01:00
Ian Barber
704b952274
Add reference to the inproc documentation
...
The previous note in connect introduced but did not elaborate on the
requirement to bind before connect in inproc. As that discussion is in
detail in the inproc docs link to there.
2012-06-12 15:49:18 +01:00
Ian Barber
7b78ed9d60
Update bind and connect documentation for clarity
...
Based on discussion with Steve O on the list, make the difference
between bind and connect usage more clear, and add a note reflecting the
fact that inproc must have bind before connect to reinforce the
information in zmq_inproc.
2012-06-12 15:46:26 +01:00
Ian Barber
e5904e63ce
Allow blocking while connect() is completing
...
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.
2012-06-12 15:34:48 +01:00
Ian Barber
409d5e8fff
Allow blocking while connect() is completing
...
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.
2012-06-12 15:31:23 +01:00
Ian Barber
95cbad3841
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 fe3fb419fe
2012-06-12 14:53:57 +01:00
Ian Barber
a563d494bb
Revert "Remove extra brackets as suggested by Martin H, and fix up a comment which was missing a word"
...
This reverts commit b79aaaf473
2012-06-12 14:53:38 +01:00
Ian Barber
cc230cfe60
Revert "And another typo on the same comment"
...
This reverts commit 297af95451
2012-06-12 14:52:38 +01:00
Ian Barber
5b167aa896
Revert "Remove the extra outpipe handling as the session is quite capable of delaying the creation of the pipe until the connection has happened. Simply don't build the pipe, and let it do that automatically."
...
This reverts commit 06485d9200
2012-06-12 14:52:18 +01:00
Ian Barber
81b8362a59
Revert "Fix a number of whitespace issues in various parts of the code, add validation to most calls on the test and take a first stab at implementing the reconnection pipe blocking."
...
This reverts commit 6f6466f088
2012-06-12 14:51:50 +01:00
Ian Barber
48d3977632
Revert "Fix condition so that PGM and EPGM sockets always create pipes immediately, even if delay_attach_on_connect is set. This allows passing through the icanhasall flag, and is realistic given the fact those protocols should be able to connect immediately"
...
This reverts commit b5ace39e2a
2012-06-12 14:51:33 +01:00
Ian Barber
6c382c5c61
Revert "Fix whitespace in test"
...
This reverts commit 0b3478d6ec
2012-06-12 14:51:19 +01:00
Ian Barber
72bae207a3
Revert "Rebinding on reconnection to allow for the pipe blocking will set the event sink again, which will cause an assert to be triggered. I have modified that to allow for setting the same sink to be OK. I believe this should be ok - on termination, if a pipe hasn't reconnected, it would just attempt to be erased from the pipes list again, which is reasonable"
...
This reverts commit 1566091bc6
2012-06-12 14:51:05 +01:00
Ian Barber
f0920caf02
Revert "On the advice of Martin Hurton, removed the new command type and just terminated the pipe in a reconnect situation, and notified the socket of the same. This handles the blocking properly, but at the cost of potentially losing in flight messages. However, this is a reasonable trade off given how much simpler it makes the patch."
...
This reverts commit c13f1d52ff
2012-06-12 14:50:50 +01:00
Ian Barber
b020bd4ba8
Revert "Remove unnecessary extra test in pipe assertion - now we are terminating pipes we don't need to allow the case in which a pipe is rebound to the same sink"
...
This reverts commit 29f8d9ca15
2012-06-12 14:49:44 +01:00
Ian Barber
956cfd9f75
Revert "Fix incorrect whitespace in if statement"
...
This reverts commit 55cbdfcf26
2012-06-12 14:49:21 +01:00
Ian Barber
74175decb4
Revert "Removing unnecessary outpipe values that had been used for reconnecting existing pipes - no longer needed when using a pipe term for the delay_attach situation."
...
This reverts commit ace7c99b91
2012-06-12 14:48:24 +01:00
Ian Barber
440af0022d
Revert "Restoring comment for clarity"
...
This reverts commit 09956dee93
2012-06-12 14:48:00 +01:00
Ian Barber
8968b294e0
Revert "When detaching a pipe, as well as checking the delay on connect sockopt is set, also ensure that the protocol is not pgm or epgm as we are not implementing the functionality for multicase types"
...
This reverts commit 5008f385ba
2012-06-12 14:47:27 +01:00
Ian Barber
836fa4b7ca
Revert "The previous code to block the socket from receiving on that pipe during a disconnect was vulnerable to a race condition. This code calls with the terminate functions of both ends of the pipe - I believe this should be safer. This required storing a pointer to the socket end of the pipe"
...
This reverts commit 336f72720a
2012-06-12 14:47:10 +01:00
Ian Barber
2c6f26153e
Revert "Reverted to a simpler shutdown. This seems to disconnect and reconnect the pipe properly, but there is a problem in overall shutdown when the pipe has blocked and reconnected - the session seems to get terminated() called on it only in shutdown for the original pipe, by which point it has been replaced. I am not sure at the moment why this only happens then, but this does mean this patch is broken at the moment"
...
This reverts commit b84b007981
2012-06-12 14:46:38 +01:00
Ian Barber
3ae68d67a4
Revert "Use the hiccup mechanism to notify the socket end of the pair of the change in state, and have it shutdown that end, and shutdown the local end normally. This seems to resolve the shutdown and race condition issues."
...
This reverts commit 67497a2643
2012-06-12 14:46:23 +01:00
Ian Barber
f5a072fee1
Revert "Code formatting, and clean up so that xhiccuped is only handled in case the pipe isn't terminating"
...
This reverts commit 5da289cd5b
2012-06-12 14:46:01 +01:00
Ian Barber
6eeaf15179
Revert "Remove unnecessary condition in assert"
...
This reverts commit 3053f7e368
2012-06-12 14:45:36 +01:00
Ian Barber
c9926f6f24
Revert "As Martin pointed out, there is a race condition in the old code where a pipe could start shutting down after disconnection, but the new one could connect first. This connection would not get a pipe created for it, so the messages could never flow. The simplest way round this would be a flag, but it is possibly for a very bouncy but fast connection to go up and down twice I imagine, so instead I have added a counter. This starts at zero, and will null out the pipe if terminate is called while it is zero. On a disconnect situation the counter is incremented, and the pipe is the not nulled if the value is non zero. In the terminated function it is decremented for each pipe that is shut down, and the assertion that the terminated pipe == the current pipe is skipped while it is non-zero. This should deal with the race condition and not allow any extra terminated() calls without hitting the assertion."
...
This reverts commit a5f7300da6
2012-06-12 14:45:14 +01:00
Ian Barber
19da88be67
Revert "Filter read and write activated calls from the pipe to the session, and delay shutdown in terminated until the final pipe is shutdown."
...
This reverts commit a90c1db7d2
2012-06-12 14:44:41 +01:00
Ian Barber
4aaf3e6389
Revert "Add disconnect test"
...
This reverts commit d8f3487ee1
2012-06-12 14:44:12 +01:00
Ian Barber
eb14890d23
Revert "Revert "Merge branch 'master' of github.com:ianbarber/libzmq""
...
This reverts commit 029d3dfae2
2012-06-12 14:43:18 +01:00
Ian Barber
bdd4e1351d
Revert "Revert "Merge branch 'master' of github.com:zeromq/libzmq""
...
This reverts commit dde69fb9fb
2012-06-12 14:28:32 +01:00
Ian Barber
029d3dfae2
Revert "Merge branch 'master' of github.com:ianbarber/libzmq"
...
This reverts commit 3345902979889b0e6f29
2012-06-12 14:13:17 +01:00
Ian Barber
dde69fb9fb
Revert "Merge branch 'master' of github.com:zeromq/libzmq"
...
This reverts commit 7b105865583345902979
2012-06-12 14:12:03 +01:00
Ian Barber
6117a2b099
Revert "Replace incomplete count with a std::set"
...
This reverts commit 4aa5ba3d11
2012-06-12 14:07:54 +01:00
Ian Barber
4aa5ba3d11
Replace incomplete count with a std::set
...
This commit removes the countdown flag and adds a set to store the pipes
that are currently being disconnected.
2012-06-12 12:13:21 +01:00
Ian Barber
7b10586558
Merge branch 'master' of github.com:zeromq/libzmq
2012-06-12 11:34:24 +01:00
Ian Barber
3345902979
Merge branch 'master' of github.com:ianbarber/libzmq
2012-06-12 11:34:17 +01:00
Ian Barber
b448de926b
Merge pull request #371 from hurtonm/fix_race_condition_in_connecters
...
Fix race conditions in {tcp,ipc}_connecter
2012-06-11 23:57:21 -07:00
Martin Hurton
3ec8e576d9
Fix race conditions in {tcp,ipc}_connecter
...
Once the object has been terminated, it is unsafe for this object
to refer to its parent.
The bug was responsible for occasional
test_shutdown_stress failures.
2012-06-12 01:47:31 +02:00
Ian Barber
d8f3487ee1
Add disconnect test
...
Add a new section to the connect delay test that disconnects and
reconnects a socket to ensure that some messages are appropriately
blocked.
2012-06-11 20:04:09 +01:00
Ian Barber
a90c1db7d2
Filter read and write activated calls from the pipe to the session, and delay shutdown in terminated until the final pipe is shutdown.
2012-06-11 07:58:59 +01:00
Ian Barber
a5f7300da6
As Martin pointed out, there is a race condition in the old code where a pipe could start shutting down after disconnection, but the new one could connect first. This connection would not get a pipe created for it, so the messages could never flow. The simplest way round this would be a flag, but it is possibly for a very bouncy but fast connection to go up and down twice I imagine, so instead I have added a counter. This starts at zero, and will null out the pipe if terminate is called while it is zero. On a disconnect situation the counter is incremented, and the pipe is the not nulled if the value is non zero. In the terminated function it is decremented for each pipe that is shut down, and the assertion that the terminated pipe == the current pipe is skipped while it is non-zero. This should deal with the race condition and not allow any extra terminated() calls without hitting the assertion.
2012-06-10 19:57:02 +01:00
Ian Barber
841cf69eb7
Merge branch 'master' of https://github.com/zeromq/libzmq
2012-06-10 19:18:18 +01:00
Ian Barber
ace8f753e1
Merge pull request #370 from hurtonm/code_cleanup
...
Use struct keyword consistently with sockaddr and sockaddr_in
2012-06-10 04:07:12 -07:00
Martin Hurton
1f53697211
Use struct keyword consistently with sockaddr and sockaddr_in
2012-06-10 02:29:49 +02:00
Pieter Hintjens
e9bfd76f87
Merge pull request #368 from hurtonm/code_cleanup
...
Small code simplification
2012-06-09 07:13:30 -07:00
Pieter Hintjens
66b7edfbd1
Merge pull request #367 from hurtonm/issue_309
...
Fix the message decoder to ignore reserved flags.
2012-06-09 07:12:48 -07:00
Martin Hurton
602c50117c
Small code simplification
2012-06-09 14:21:01 +02:00
Martin Hurton
796212a985
Fix the message decoder to ignore reserved flags.
...
Failing to clear the reserved flags, the decoder may produce
messages with 'identity' and 'shared' flags set.
This unintended modification of message flags can lead to memory
errors or asserion failures.
Fixes issue #309
2012-06-09 13:39:44 +02:00
Ian Barber
3053f7e368
Remove unnecessary condition in assert
2012-06-09 00:08:14 +01:00
