Commit message (Collapse) | Author | Age | Files | Lines | |
---|---|---|---|---|---|
* | refactoring: rename "round robin" to "active consumers"bug20943 | Matthias Radestock | 2009-06-10 | 1 | -27/+31 |
| | |||||
* | proper round-robining even when blocking | Matthias Radestock | 2009-06-10 | 1 | -124/+111 |
| | | | | | | | | | | | | | | | | | | | | This involved some substantial changes to the queue internal data structures - mostly by choice; the new design is cleaner: - We no longer keep a list of consumers in the channel records. Now the channel records just contain a consumer count instead, and that's only there for efficiency so we can more easily tell when we need to register/unregister with the limiter. - We now keep *two* consumer queues - one of active consumers (that's the one we've always had) and one of blocked consumers. We round-robin on the first one as before, and move things between the two queues when blocking/unblocking channels. When doing so the relative order of a channel's consumers is preserved, so the effects of any round-robining the active consumers get carried through to the blocked consumers when they get blocked and then back to the active consumers when they get unblocked. | ||||
* | emergency merge of bug20958 into default | Matthias Radestock | 2009-06-10 | 3 | -14/+21 |
|\ | | | | | | | | | | | We point to the macports files of the default branch from our web site and they got broken with the merge of bug20333. This hopefully fixes that, but further qa is required. | ||||
| * | merge in bug20955 | Matthew Sackman | 2009-06-10 | 1 | -4/+11 |
| |\ | |||||
| | * | respect channel blocking when adding consumersbug20955 | Matthias Radestock | 2009-06-10 | 1 | -4/+11 |
| |/ | |||||
| * | Merge bug20801 into default | Hubert Plociniczak | 2009-06-10 | 1 | -8/+10 |
| |\ | |||||
| | * | Merge two default branches into one. | Hubert Plociniczak | 2009-06-10 | 3 | -23/+50 |
| | |\ | |||||
| | * | | fix off-by-one error | Matthias Radestock | 2009-06-10 | 1 | -1/+1 |
| | | | | | | | | | | | | | | | | not that it really matters much | ||||
| | * | | cosmetic | Matthias Radestock | 2009-06-10 | 1 | -7/+9 |
| | | | | |||||
| * | | | Use dist macro for building correct distribution versions.bug20801 | Hubert Plociniczak | 2009-06-10 | 2 | -5/+3 |
| | |/ | |/| | |||||
* | | | Removed default shell, changed wrapper to work without default shell | timc | 2009-06-10 | 2 | -2/+2 |
| | | | |||||
* | | | Creates rabbitmq user with a default shell and uses correct arguments to the ↵ | timc | 2009-06-10 | 2 | -3/+3 |
|/ / | | | | | | | os x su command | ||||
* | | Merge from bug20333 post verification | timc | 2009-06-10 | 3 | -23/+50 |
|\ \ | |/ |/| | |||||
| * | remove/add files missed on previous commitbug20333 | Matthias Radestock | 2009-04-06 | 2 | -2/+23 |
| | | |||||
| * | use wrapper scripts in macports | Matthias Radestock | 2009-04-06 | 1 | -21/+27 |
| | | |||||
* | | merge bug20940 into default | Matthias Radestock | 2009-06-10 | 2 | -9/+12 |
|\ \ | |||||
| * | | cosmeticbug20940 | Matthias Radestock | 2009-06-10 | 1 | -6/+7 |
| | | | |||||
| * | | Made can_send/2 into can_send/3. | Matthew Sackman | 2009-06-09 | 2 | -9/+11 |
|/ / | |||||
* | | merging in bug20782 | Matthew Sackman | 2009-06-08 | 8 | -81/+103 |
|\ \ | |||||
| * | | clean up tx records in queues when a transaction's channel diesbug20782 | Matthias Radestock | 2009-06-05 | 4 | -23/+43 |
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Previously queues were only monitoring channels with subscribers or to which ack-requiring messages had been delivered. Now queues also monitor channels from which they have received transactional publishes. Queues record the last tx id they have seen from a channel. This then makes it easy and efficient to find the associated tx record in the queue's process dictionary when a channel process dies - the alternative would be to scan the tx records for matching channel pids - and perform the required rollback activities for the tx. | ||||
| * | | refactoring: bundle up all the data for a publish | Matthias Radestock | 2009-06-05 | 7 | -64/+66 |
|/ / | | | | | | | | | Passing this around as separate args was becoming a pain. Also, now it's easier to add more data items. | ||||
* | | Split out internal_declare/2. | Tony Garnock-Jones | 2009-06-04 | 1 | -2/+10 |
| | | |||||
* | | merge bug20633 into default | Matthew Sackman | 2009-06-04 | 1 | -16/+24 |
|\ \ | |||||
| * | | make delete_binding return an error when the binding does not existbug20633 | Matthias Radestock | 2009-06-02 | 1 | -16/+24 |
| | | | | | | | | | | | | | | | This required some refactoring in order to avoid duplication of the code that constructs #binding records. | ||||
* | | | merge bug20877 into default | Matthias Radestock | 2009-06-03 | 0 | -0/+0 |
|\ \ \ | |||||
* | | | | alignbug20877 | Matthias Radestock | 2009-06-03 | 1 | -2/+2 |
| | | | | |||||
* | | | | Change from number to atom. | Tony Garnock-Jones | 2009-06-03 | 2 | -5/+11 |
|/ / / | |||||
* | | | merge bug20654 into default | Tony Garnock-Jones | 2009-06-03 | 7 | -86/+178 |
|\ \ \ | |||||
| * \ \ | mergerbug20654 | Tony Garnock-Jones | 2009-06-03 | 1 | -1/+1 |
| |\ \ \ | |||||
| | * | | | fix typo | Matthias Radestock | 2009-05-29 | 1 | -1/+1 |
| | | | | | |||||
| * | | | | merge default into bug20654 | Tony Garnock-Jones | 2009-05-27 | 14 | -54/+77 |
| |\ \ \ \ | |/ / / / |/| / / / | |/ / / | |||||
* | | | | merge v1_5 into default | Tony Garnock-Jones | 2009-05-27 | 4 | -10/+19 |
|\ \ \ \ | |_|/ / |/| | | | |||||
| * | | | Added tag rabbitmq_v1_5_5 for changeset 5dc463f43bb7v1_5 | Matthias Radestock | 2009-05-19 | 0 | -0/+0 |
| | | | | |||||
| * | | | add package changelog entries for 1.5.5 releaserabbitmq_v1_5_5 | Matthias Radestock | 2009-05-19 | 3 | -1/+10 |
| | | | | |||||
| * | | | merge bug20676 into v1_5 | David Wragg | 2009-05-18 | 3 | -10/+17 |
| |\ \ \ | |||||
| * | | | | Change to dependenciesbug20676 | Matthew Sackman | 2009-05-18 | 1 | -1/+1 |
| | | | | | |||||
* | | | | | helper method for not_found errors | Matthias Radestock | 2009-05-20 | 4 | -12/+9 |
| | | | | | | | | | | | | | | | | | | | | There was enough repetition to make it worthwhile | ||||
* | | | | | document support for headers exchange | Matthias Radestock | 2009-05-19 | 1 | -1/+1 |
| | | | | | | | | | | | | | | | | | | | | We forgot to do this bug 20152. | ||||
* | | | | | merge v1_5 into default | Tony Garnock-Jones | 2009-05-18 | 34 | -777/+2376 |
|\ \ \ \ \ | |_|/ / / |/| | | | | |||||
| * | | | | merge v1_5 into default | Tony Garnock-Jones | 2009-05-13 | 1 | -1/+5 |
| |\ \ \ \ | |||||
| * \ \ \ \ | merge v1_5 into default | Tony Garnock-Jones | 2009-05-13 | 34 | -777/+2376 |
| |\ \ \ \ \ | |||||
* | \ \ \ \ \ | merge bug20767 into v1_5 | Tony Garnock-Jones | 2009-05-18 | 3 | -10/+17 |
|\ \ \ \ \ \ \ | |_|_|_|/ / / |/| | | | | | | |||||
| * | | | | | | Silence unused variable warningbug20767 | Tony Garnock-Jones | 2009-05-18 | 1 | -1/+1 |
| | | | | | | | |||||
| * | | | | | | display node name and db dir on startup | Matthias Radestock | 2009-05-18 | 3 | -10/+17 |
|/ / / / / / | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | ...in a pretty and easy-to-parse way In the process we introduce rabbit_mnesia:dir() so we don't depend directly on the mnesia module - this also removes a dependency in the persister. | ||||
* | | | | | | Merge bug20745 into v1_5 | Hubert Plociniczak | 2009-05-16 | 1 | -0/+2 |
|\ \ \ \ \ \ | |||||
| * \ \ \ \ \ | Merge bug20679 into v1_5 | Hubert Plociniczak | 2009-05-16 | 1 | -0/+2 |
| |\ \ \ \ \ \ | | |_|_|/ / / | |/| | | | | | |||||
| | * | | | | | Add Default-Start and Default-Stop LSB init script linesbug20679 | David Wragg | 2009-05-08 | 1 | -0/+2 |
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | To shut rpmlint up. | ||||
* | | | | | | | Don't package the INSTALL documentbug20745 | David Wragg | 2009-05-13 | 1 | -1/+1 |
|/ / / / / / | |||||
* | | | | | | merge bug20702 into v1_5 | Tony Garnock-Jones | 2009-05-13 | 1 | -1/+5 |
|\ \ \ \ \ \ | |_|/ / / / |/| | | | | | |||||
| * | | | | | Set PATH to workaround case when escript is not in /usr/binbug20702 | David Wragg | 2009-05-08 | 1 | -1/+5 |
| | | | | | |