c2a3107672
Move max_hops to upstream_set.
2011-06-17 13:19:55 +01:00
54ffd130f8
Fix a list vs binary bug around default exchanges / vhosts.
2011-06-17 12:54:46 +01:00
b7e7019065
Remove explicit specification of the exchange
2011-06-17 12:54:24 +01:00
c639bf7385
If we log disconnections / failed connections, we should probably log successful connections too.
2011-06-17 12:40:16 +01:00
0beb5c72c6
If the upstream channel goes down for an intelligible reason, just log it and die quietly.
2011-06-17 12:34:36 +01:00
6943d4bf37
Rephrase
2011-06-17 12:04:38 +01:00
e8db3d7ae3
Update the warnings / limitations.
2011-06-17 11:58:02 +01:00
e626a81e89
Less of an explosion when the other end cannot be contacted.
2011-06-16 18:26:02 +01:00
aa4bb71de4
If the downstream channel is going down cleanly, don't blow up, we're likely to go down very soon anyway.
2011-06-16 18:19:23 +01:00
54497b3e0f
Test max_hops by building a ring with three brokers.
2011-06-16 17:56:23 +01:00
ebd5185e94
Check length of AMQP array properly.
2011-06-16 17:55:59 +01:00
4e3c2410e3
Acknowledge dropped messages, or they'll build up forever, oops.
2011-06-16 17:55:36 +01:00
fbeec700e9
Ignore app file properly.
2011-06-16 16:16:23 +01:00
bda7f423a8
Simplistic max_hops. Needs testing.
2011-06-16 15:44:34 +01:00
329fc5a787
Remove purpose argument, let bindings propagate freely.
2011-06-16 14:48:06 +01:00
18fbdfb62e
Rename x-forwarding to x-received-from, remove protocol from it.
2011-06-16 13:23:36 +01:00
c4b6c10a64
Essay.
2011-06-07 14:29:10 +01:00
de9af501f1
Essay still to come, but here's a couple of comments in place to explain some bits.
2011-06-07 13:26:49 +01:00
767555d9fc
Another attempt at explaining these, by explictly discussing how it works. Also remove the (* 1000) thing, that doesn't make things any clearer.
2011-06-06 13:28:58 +01:00
4c2d21f879
Attempt to make docs clearer.
2011-06-06 13:07:10 +01:00
942601b847
Update documentation for that.
2011-06-06 12:39:22 +01:00
b9ba315033
Rename queue_expires to expires, add support for message-ttl.
2011-06-06 12:17:12 +01:00
153be18672
Rejig the configuration to go through an additional layer of indirection so that exchange arguments can plausibly be immutable, and to get rid of the confusing merging.
...
Also fix the insidious default queue expiry that had been hiding.
2011-06-03 18:09:23 +01:00
e1f297f875
Make all federation objects have a prefix to their name, generate a more unique nodename and allow the admin to override it.
2011-06-02 18:28:05 +01:00
cf06cfd19a
Remove the special handling of serial being undefined, and explain why we don't care.
2011-06-02 17:55:14 +01:00
ff943bbc75
Don't delete the upstream Q on reconnect if the downstream X is transient. I had silly ideas that that corresponded more to the semantics, but not really.
2011-06-02 17:38:34 +01:00
6c71bcf216
If we find the serial is not defined we don't actually want to blow up, this could be normal, we're just a short-lived exchange. So instead go into a mode where the only thing we can do is stop.
2011-06-02 17:36:53 +01:00
f8c6b0a0c2
One more instance.
2011-06-02 15:21:24 +01:00
9f0961b181
Oops
2011-06-02 15:15:28 +01:00
ec1bfd11e0
Move pget / set_table_value to rabbit_misc
2011-06-02 15:05:05 +01:00
8a96385b04
"Handle" getting not_found from peek_serial (not that we can do anything, it's pretty much a sign that we're dying anyway, so lets die) + cosmetic.
2011-06-02 14:31:34 +01:00
98b0d038e9
refactor: inline and re-arrange
2011-06-01 20:43:47 +01:00
6a2816ea99
some refactoring and cosmetics:
...
- fewer intermediate vars
- more logical sequence of operations on connection establishment
2011-06-01 19:15:25 +01:00
04e83ea49a
Fix the command-queuing logic. This requires a number of changes:
...
* Actually fix the logic: keep track of the next expected
serial. Queue up commands that come after it, and ignore ones
before it. Take initial value of the serial in the same tx as
we execute rabbit_binding:list_for_source.
* Introduce a function in rabbitmq-server to peek the next value
of an exchange's serial to support this.
* Make (un)binding transmission asynchronous, and note it in the
README. This is the right thing to do, but unfortunately it
means we need to insert sleep()s all over the place in the
test suite. Ah well.
* Make remove_bindings into a single message sent to each link,
rather than a bunch of (remove_binding)s.
* Make add_bindings always send a message to each link, even if
it does nothing, to eat up the serial number.
* Make the serialisation unit test actually declare exchanges.
2011-06-01 16:38:39 +01:00
1784f7dfa4
Oops.
2011-05-31 13:51:19 +01:00
4cd0c697e1
Add a (currently stupid) test to ensure we serialise the add / remove bindings calls properly.
2011-05-31 13:51:12 +01:00
e7c2e90253
Move #upstream{}-munging into rabbit_federation_upstream. This increases the amount of pget copies in the world, but we'll deal with that soon.
2011-05-27 17:35:20 +01:00
5f9f2efa39
cosmetic
2011-05-27 14:50:27 +01:00
4e00092165
cosmetic
2011-05-27 14:36:01 +01:00
06b9b71e62
Rearrange exports
2011-05-27 14:16:52 +01:00
2e1de9fe2f
Post bug 24079 update
2011-05-27 13:57:30 +01:00
24f1a4017a
Various simple / uncontroversial cleanups.
2011-05-27 13:54:05 +01:00
80208e86f0
Merge heads
2011-05-27 13:19:18 +01:00
faff573a89
This is not great, but necessary to run the tests on something other than an Optiplex. Still faster than many other test suites...
2011-05-27 11:57:55 +01:00
2a13b663d7
merge heads
2011-05-27 06:21:19 +01:00
f336da1c49
renaming refactor
2011-05-26 13:57:41 +01:00
79167d6c76
cosmetic
2011-05-26 13:55:13 +01:00
486600bb92
Various clarifications + list of enhancements.
2011-05-26 12:15:14 +01:00
d4c060a586
cosmetics
...
- write funs as a single line when we can
- define funs in the same order as the exports
- only use section marker comments for clearly delineated sections,
and keep them quite coarse grained
- other minor tweaks
2011-05-26 09:57:41 +01:00
26e0c6d625
Clarify provenance of and changes in pg2_fixed
2011-05-26 09:17:01 +01:00
25a254da12
less will
2011-05-26 08:04:41 +01:00
48621877bb
Typo
2011-05-19 13:17:33 +01:00
f6ae916866
Update now that things have been merged to default
2011-05-17 14:44:00 +01:00
5adf0b4429
This looked like an error and confused me.
2011-05-04 16:09:44 +01:00
5027f1f564
App name has changed
2011-05-04 16:02:27 +01:00
6b6c32bf42
Application names: rabbit_foo -> rabbitmq_foo
2011-04-13 16:02:31 +01:00
87e333be56
Build system changes.
2011-04-12 12:18:42 +01:00
b903cd07b0
Merge bug 23628 and update for API changes in bug 23939.
2011-04-08 17:05:38 +01:00
804299b424
API change
2011-04-01 17:40:46 +01:00
38374291ad
Cosmetic.
2011-04-01 14:14:29 +01:00
ee319f74fc
Revert a605a7bde03c, of course that doesn't make sense, what if the link restarts, the bindings will be wrong.
2011-04-01 13:52:16 +01:00
2fb9d57f18
Thread through bindings from start callback, get rid of the last use of rabbit_binding:list_for_source.
2011-04-01 12:21:18 +01:00
c770696ab6
API changes.
2011-04-01 12:05:20 +01:00
a288f25ee1
updating use of amqp_client exit reasons
2011-03-29 17:41:43 +01:00
5e29f9392a
Slightly more doc.
2011-03-29 11:52:17 +01:00
5c6900f1d1
API change.
2011-03-29 11:45:55 +01:00
f4fd3ac4be
API change.
2011-03-28 14:18:46 +01:00
5f36fa6d70
Fix link startup. Tests all pass again.
2011-03-28 13:17:31 +01:00
1e38436b14
Use process groups rather than navigating the supervisor hierarchy to talk to exchange pid sets. The restart test fails, but it's Friday night.
2011-03-25 18:27:30 +00:00
85e2ab33fd
Create a links API.
2011-03-24 12:46:20 +00:00
6ae679b635
bug23939 API changes.
2011-03-23 14:14:02 +00:00
6a31956637
Symmetry.
2011-03-23 12:33:06 +00:00
ad8428d9fc
This makes tracking down bugs rather easier.
2011-03-23 12:26:14 +00:00
1081f99586
Split up smart_unbind_test and find a way to test that unbinds get sent upstream.
2011-03-23 11:43:43 +00:00
8f5bf686b8
A bit more general and hopefully clearer.
2011-03-23 11:33:24 +00:00
8cb922f1c2
Refactor tests a bit.
2011-03-23 11:11:26 +00:00
b8e63691c4
Somewhat embarassingly that's been wrong for a while. But I'm not sure what test we could use to check it that doesn't e.g. look directly at the upstream bindings. Ugh.
2011-03-22 18:01:30 +00:00
11c8fa7b9a
The previous commit was buggy because we never added bindings. Add more test to catch that, fix it, and optimise add_binding as well.
2011-03-22 16:35:25 +00:00
813e4d1431
Sadly that little badmatch was inaudible due to the continuous barrage of explosions.
2011-03-22 16:17:11 +00:00
59d449e7bf
Reduce use of rabbit_binding:list_for_source/1 by keeping track ourselves of where we're at.
2011-03-22 16:05:49 +00:00
75f6e1c4a2
serialise_events/1.
2011-03-21 12:28:29 +00:00
2163ab4831
Simplify.
2011-03-21 11:47:30 +00:00
3cd4b78949
Rename rabbit_federation_exchange_upstream to rabbit_federation_link.
2011-03-21 11:41:57 +00:00
92a65e37c2
Slightly smaller, quite a lot more correct.
2011-03-21 11:31:26 +00:00
75b2cad486
Ensure that we use Mnesia's ordering for add / remove bindings.
2011-03-18 17:44:50 +00:00
deada309ce
Remove maybe_unbind_upstreams and push its logic into the exchange processes
2011-03-18 17:16:46 +00:00
0923020d4f
Make sure rabbit_federation_db:forget_exchange/1 actually gets called.
2011-03-04 18:15:00 +00:00
18bb7e00ed
That seemed to break every logging function. Pity.
2011-03-04 17:58:16 +00:00
b337ff2ed0
Don't keep durability in the state forever, we mostly don't care.
2011-03-04 11:55:10 +00:00
9fda27efb5
Cosmetic + comments.
2011-03-04 11:10:47 +00:00
96ab34354e
Improve documentation again.
2011-03-04 10:43:12 +00:00
6706a27004
Allow port to be any numeric type.
2011-03-04 10:43:03 +00:00
f5d3229a32
Remove TODO, the issues have gone into bugzilla.
2011-03-03 17:50:05 +00:00
63ba1f7164
Default to no prefetch count.
2011-03-03 17:29:57 +00:00
78f2c255f1
So, err, the delivery tag map is no longer needed now that we have a one-one channel relationship.
2011-03-03 16:48:29 +00:00
49420a7930
If we blow up while opening connections and channels, make sure we close the ones that got opened.
2011-03-03 16:30:27 +00:00
4fd555e2c9
Name the file properly to prevent DETS auto-repair from blowing up, but handle the error if it does anyway.
2011-03-03 15:42:06 +00:00
08caf247fc
Clean up upstream_from_table/2 substantially.
2011-03-03 15:29:00 +00:00
56bc814d8c
Remove URIs from configuration.
2011-03-03 14:13:05 +00:00
c8c974552e
Add protocol, fix host.
2011-03-03 12:16:47 +00:00
6ff9b80cc4
Further pushback of URIs.
2011-03-03 12:05:20 +00:00
b4fb8e6f18
Remove URIs from exchange_upstream.
2011-03-03 12:01:17 +00:00
6d1b4e1a60
Make tests independent of any crazy configuration file you may have lying around.
2011-03-03 10:43:47 +00:00
2debec04a5
Oops, forgot to add this...
2011-03-03 10:33:31 +00:00
afe72dfc0e
It's been a long day.
2011-03-02 17:16:38 +00:00
6fb49364ff
Handle non-guest connections locally.
2011-03-02 17:13:29 +00:00
c5136d8538
Damn, I thought of something else.
2011-03-02 16:37:23 +00:00
eff6725109
Configurable retry interval / queue expiry / prefetch count / heartbeat.
2011-03-02 16:29:21 +00:00
5363229765
Use uri_parser that is now in the broker.
2011-03-02 15:09:25 +00:00
da406e466c
Rename uri_parser as a simple way to avoid clashing with the one in rabbitmq-shovel.
2011-03-02 14:46:45 +00:00
6db99c1f52
TODO and WARNING
2011-03-02 14:39:39 +00:00
235223e334
Shorter names, determine if we created something by existence of a special argument.
2011-03-02 14:31:53 +00:00
693b9b800e
Reset bindings on connect.
2011-03-02 13:13:30 +00:00
daa1d7dbe7
Use DETS, abstract a bit.
2011-03-01 16:09:34 +00:00
2a7945efec
More information.
2011-03-01 13:18:13 +00:00
55d3aa399b
Tweak.
2011-03-01 13:13:53 +00:00
65f899be4d
Tweak.
2011-03-01 13:12:15 +00:00
c395ef1c2b
More doc, focussing on "why" rather than "how".
2011-03-01 13:09:33 +00:00
f49355e047
Documentation.
2011-03-01 12:50:42 +00:00
ffe8ed1074
Deal with downstream channel going down, it's trivial now.
2011-03-01 11:04:00 +00:00
49e035243d
Fix various bugs:
...
Make supervision more correct
Make shutdown work
Fix durability in tests
2011-03-01 11:01:42 +00:00
c3031441b1
Split up such that we have one process per (upstream, downstream) combination. This mostly works, but there's something weird with the supervision hierarchy keeping the Erlang VM alive when rabbit_federation_exchange_upstream is not temporary.
2011-02-28 18:10:01 +00:00
119fc3e04f
Don't exclude all those non-RabbitMQ brokers supporting AMQP 0-9-1, SASL EXTERNAL and (soon) e2e bindings.
2011-02-24 13:23:10 +00:00
944cd4993b
Support connections with alternate creds, SSL, EXTERNAL.
2011-02-24 13:19:00 +00:00
86d41ae23e
Create the exchange process at exchange recovery. This means moving the processes / supervisor under the rabbit application, which means making them be able to run in a dumb state when the amqp_client is not yet available, which means splitting code out into a bunch of modules, etc etc.
2011-02-23 16:47:50 +00:00
7a9b3140a8
Bah
2011-02-22 17:41:45 +00:00
599ed44559
We only ever use exchange from the upstream properties.
2011-02-22 17:26:35 +00:00
1cdaeef21f
Use rabbitmqctl wait
2011-02-22 17:02:15 +00:00
24479f9ba7
TODO
2011-02-22 17:02:02 +00:00
6a9a6e896e
Don't allow binding, unbinding or publishing loops, by the simple expedient of only allowing things to cross one hop.
2011-02-22 16:29:42 +00:00
6c83263973
TODO
2011-02-18 17:19:00 +00:00
5f0cd4d884
Add forwarding information in a header.
2011-02-18 16:27:12 +00:00
40b80e9f8a
Only transmit unbinds upstream if it's the last with the same {Source, Key, Args}.
2011-02-18 13:40:10 +00:00
cdb9c1e20d
In multiple_downstreams_test messages can arrive in any order. Deal with that and stop being racy.
2011-02-18 12:51:51 +00:00
45f9aad9f1
Conserve vertical space.
2011-02-18 12:46:31 +00:00
2efd415f87
Delete upstream queues when exchange is deleted. Remove childspec from supervisor when exchange is stopped.
2011-02-18 11:59:20 +00:00
6e4d1f6d81
Validate args
2011-02-17 18:14:06 +00:00
3bc063c990
Another use of with_disposable_channel.
2011-02-17 17:10:43 +00:00
a3bcddc176
Delete and recreate upstream queue when creating transient exchange
2011-02-17 17:08:08 +00:00
f756188eac
Make the static configuration more useful / flexible.
2011-02-17 16:30:13 +00:00
fe192c1b78
TODO
2011-02-17 15:40:22 +00:00
77fdde4bef
TODO
2011-02-17 15:37:17 +00:00
e74bd6b923
Use acks / confirms to make sure we're never responsible for anything.
2011-02-17 15:23:48 +00:00
6c9e11df7c
Turns out e2e just works, but let's prove that (and note that we're not completely efficient here, we send all msgs that get routed to the exchanges bound to the federation exchange, regardless of whether they get routed anywhere after that).
2011-02-17 12:49:26 +00:00
1697acd3d1
TODO
2011-02-16 12:14:43 +00:00
d85ca4a125
That's a binary now.
2011-02-15 18:26:50 +00:00
5e573553ee
Handle the upstream broker going down.
2011-02-15 18:25:21 +00:00
c1cf725a5d
Grab the port number from URLs, add a test across two brokers.
2011-02-15 14:30:42 +00:00
8c0b0118a6
Another attempt at readable names. And no, this should not be our own URI, we don't know that.
2011-02-15 13:28:33 +00:00
690eb40534
Revert "readable name" for the time being as it's not unique enough. And add the test that let me find that.
2011-02-15 12:28:35 +00:00
011c19dfe1
Separate out the config based exchange tests entirely.
2011-02-15 12:15:23 +00:00
66c2bb1d85
Don't store the backing module in the ETS table, we already have it in the args. This means we always have it, and can always call with_module(...)
2011-02-15 12:15:06 +00:00
d2ae33229c
Use a readable queue name, and don't have it expire immediately.
2011-02-11 18:26:00 +00:00
2b875b4c48
Improvements to shutdown.
2011-02-11 17:17:16 +00:00
f41265b56a
TODO
2011-02-11 14:46:06 +00:00
537f7b9c69
Multiple upstreams.
2011-02-11 14:44:37 +00:00
1c787900eb
Be more resilient when unbinding.
2011-02-11 13:45:51 +00:00
3b2c67075e
Refactor tests, test configuration, allow more than one federation exchange per broker(!)
2011-02-11 13:33:32 +00:00
e8bd411781
The simplest possible test.
2011-02-11 12:06:27 +00:00
ffe698a346
TODO
2011-02-11 11:23:43 +00:00
6c8101b8c0
Make some attempt to assert args equivalence and remove bindings.
2011-02-10 18:16:44 +00:00
6371fa1cd9
Rename local / remote to downstream / upstream.
2011-02-10 17:59:33 +00:00
438df4c513
TODO
2011-02-10 17:49:01 +00:00
da339a0513
Give the federation exchange another type of exchange to emulate. This sorta works now (we can federate a message! woo!), but in a couple of cases we need the backing module before we have it. Hmm.
2011-02-10 17:11:33 +00:00
8f4cf84e23
The beginnings of a federation plugin. Does not work.
2011-02-10 16:26:32 +00:00
Simon MacMullen
Matthias Radestock