summaryrefslogtreecommitdiff
path: root/src/rabbit_channel.erl
Commit message (Collapse)AuthorAgeFilesLines
* Bring back impersonator tag.bug26098Simon MacMullen2014-04-041-4/+8
|
* Monitor consuming queue regardless of whether the consumer is CCN capable, ↵bug26070Simon MacMullen2014-03-251-21/+17
| | | | only decide whether to send the CCN on that basis.
* Our other consumer arg has an x- on it. Also, let's validate.Simon MacMullen2014-03-251-1/+1
|
* Swap the default value for cancel-on-ha-failover.Simon MacMullen2014-03-251-2/+2
|
* Add a comment, and rename the thing to correspond more with words we use ↵Simon MacMullen2014-03-211-4/+9
| | | | already.
* rabbit_amqqueue:basic_consume() will wait for recovery anyway, we don't need ↵Simon MacMullen2014-03-211-10/+7
| | | | a special function for that.
* Attempt at error checking the basic.consume.Simon MacMullen2014-03-201-22/+31
|
* Abstract the waiting into rabbit_amqqueue.Simon MacMullen2014-03-201-10/+5
|
* Rename variable (and catch a case I missed before).Simon MacMullen2014-03-201-8/+8
|
* OopsSimon MacMullen2014-03-191-1/+1
|
* First hack at recovering consumers.Simon MacMullen2014-03-191-60/+88
|
* Update copyright for 2014bug25940Simon MacMullen2014-03-171-1/+1
|
* Rename info itemsSimon MacMullen2014-03-061-3/+3
|
* Change the semantics of the basic.qos global flag to switch between ↵Simon MacMullen2014-03-041-10/+19
| | | | per-consumer and per-channel prefetch.
* Merge in defaultSimon MacMullen2014-02-101-10/+10
|\
| * Get rid of the impersonator tag, only let the magic user do it.Simon MacMullen2014-02-051-8/+7
| |
| * cosmeticMatthias Radestock2014-02-031-3/+3
| |
| * Merge bug24408 (again)Simon MacMullen2014-02-031-88/+21
| |\
| * | This optimisation only makes sense when the message is neither mandatory nor ↵bug24408Simon MacMullen2014-02-031-0/+1
| | | | | | | | | | | | confirmable!
* | | Include a reference in forced event refreshes.Simon MacMullen2014-02-101-6/+7
| |/ |/|
* | Merge bug 25817Simon MacMullen2014-01-241-187/+138
|\ \
| * \ merge bug24408 into defaultMatthias Radestock2014-01-231-41/+60
| |\ \ | | |/
| | * some drive-by inliningMatthias Radestock2014-01-221-8/+3
| | | | | | | | | | | | not strictly part of this bug but easier here than on default
| | * some inlining, cosmetics, and more explicit sequencingMatthias Radestock2014-01-221-22/+20
| | |
| | * s/confirmed/confirm/g, fix a test.Simon MacMullen2014-01-221-2/+2
| | |
| | * Remove some duplication.Simon MacMullen2014-01-221-7/+7
| | |
| | * Don't stomp on timeout when receiving mandatory notification.Simon MacMullen2014-01-221-2/+3
| | |
| | * Fix basic.return stats, and fix various problems found by dialyzer.Simon MacMullen2014-01-221-14/+4
| | |
| | * OopsSimon MacMullen2014-01-221-2/+0
| | |
| | * Substantially improve both performance and correctness: 1) really don't ↵Simon MacMullen2014-01-221-23/+29
| | | | | | | | | | | | treat presence of MsgSeqNo as a cue we are doing confirms, and 2) get the right dtree function when we receive a MsgSeqNo.
| | * Get dtree use right: we want take_all/2 when we get a mandatory ack (since ↵Simon MacMullen2014-01-211-7/+5
| | | | | | | | | | | | one queue getting it is enough). take/2 on DOWN is correct.
| | * Quick-and-dirty version of pipelined mandatory publishingSimon MacMullen2014-01-211-21/+54
| | |
| * | refactor: make 'reject' arg order consistent with 'ack' et alMatthias Radestock2014-01-221-1/+1
| | | | | | | | | | | | cherry-picked from bug24297
| * | refactor: basic_consume/10 -> /9Matthias Radestock2014-01-221-13/+1
| | | | | | | | | | | | cherry-picked from bug24297
| * | inlineMatthias Radestock2014-01-221-11/+7
| | |
| * | merge bug25962 into defaultMatthias Radestock2014-01-211-9/+14
| |\ \ | | |/
| | * cosmeticMatthias Radestock2014-01-211-9/+5
| | |
| | * stable to defaultSimon MacMullen2014-01-171-0/+9
| | |\
| | | * Oops x 2bug25965Simon MacMullen2014-01-171-1/+1
| | | |
| | | * Limit maximum message size.Simon MacMullen2014-01-171-0/+9
| | | |
| * | | eliminate unused var warningbug25962Matthias Radestock2014-01-151-1/+1
| | | |
| * | | drop support for channel.flow{active=false}Matthias Radestock2014-01-151-60/+8
| |/ /
| * | tiny simplifying refactorMatthias Radestock2014-01-101-2/+2
| | |
| * | Revert changes to add proc_name via gen_server2, instead do it explicitly ↵bug25910Simon MacMullen2014-01-101-2/+2
| | | | | | | | | | | | with a helper macro.
| * | Merge in defaultSimon MacMullen2014-01-101-1/+2
| |\ \
| | * | Add queue_coordinator / queue_syncer / gm. Shift to getting gen_server2 to ↵Simon MacMullen2013-12-101-2/+2
| | | | | | | | | | | | | | | | store the name in most cases.
| | * | RenameSimon MacMullen2013-12-101-1/+1
| | | |
| | * | A bit more abstraction, plus identities for heartbeater /Simon MacMullen2013-12-101-1/+1
| | | | | | | | | | | | | | | | queue_collector / limiter.
| | * | Store some process identity in the process dictionary.Simon MacMullen2013-12-051-0/+1
| | | |
| * | | Merge bug25817Simon MacMullen2014-01-101-54/+48
| |\ \ \
View Lorry Controller Status