summaryrefslogtreecommitdiff
Commit message (Expand)AuthorAgeFilesLines
* Removed some transactions and made all transaction bodies idempotent. They we...bug20980Matthew Sackman2009-08-051-74/+56
* merge in from 21087Matthew Sackman2009-08-0322-931/+4582
|\
| * just making set_mode as a pcast in the disk_queue just like it is in the amqq...Matthew Sackman2009-08-011-1/+1
| * merging in from 21087. All tests pass.Matthew Sackman2009-07-293-96/+163
| |\
| * | Prefetcher appears to be done and working well. None of the tests exercise it...Matthew Sackman2009-07-214-97/+262
| * | sigh, another stupid bug which none of the tests catch and which I just happe...Matthew Sackman2009-07-211-2/+2
| * | Fixed a bug in the mixed_queue which could lead to messages being marked unde...Matthew Sackman2009-07-214-19/+29
| * | minor doc typeos and formattingMatthew Sackman2009-07-211-12/+13
| * | bare non-functioning skeleton of prefetcher. Essay written on design of prefe...Matthew Sackman2009-07-213-2/+206
| * | Stripping out old broken prefetch. Also reverted gen_server2 back to the revi...Matthew Sackman2009-07-213-254/+83
| * | Fixed the commit bug. Really this should probably be in bug20470 but I really...Matthew Sackman2009-07-191-51/+48
| * | Spotted and corrected some mistakes where messages published to the mixed que...Matthew Sackman2009-07-192-7/+7
| * | ok, limits on the cache, and on prefetch.Matthew Sackman2009-07-171-18/+30
| * | part 2 done. The mixed_queue is back to using only one queue. Start up time i...Matthew Sackman2009-07-173-152/+114
| * | The use of the in-memory run length queue in disk_only queue is considered a ...Matthew Sackman2009-07-173-67/+98
| * | Well it's better. The memory size is now recovered at start up by doing a fol...Matthew Sackman2009-07-165-73/+94
| * | Substantial changes to mixed_queue.Matthew Sackman2009-07-152-224/+264
| * | Just adding a bit more testing really just to bump the code coverage up.Matthew Sackman2009-07-102-1/+46
| * | prefetch part 1. When a mixed_queue sees that the next item in its queue is o...Matthew Sackman2009-07-103-40/+85
| * | Adjusted documentationMatthew Sackman2009-07-101-7/+11
| * | *cough*Matthew Sackman2009-07-091-1/+2
| * | ...and with some testing and debugging, it might even work as described in th...Matthew Sackman2009-07-092-28/+24
| * | additional documentationMatthew Sackman2009-07-091-1/+9
| * | Initial work to permit low priority background tasks to be catered for.Matthew Sackman2009-07-095-78/+236
| * | merging in from 21087. In testing, observed oscillation - with lots of queues...Matthew Sackman2009-07-092-16/+17
| |\ \
| * | | length is never used in disk_queue, so removedMatthew Sackman2009-07-091-21/+12
| * | | minor documentation fix.Matthew Sackman2009-07-091-2/+1
| * | | Fixes from removing the non-contiguous sequences support from the disk queue ...Matthew Sackman2009-07-091-31/+20
| * | | The mixed queue contains in its queue knowledge of whether the next message i...Matthew Sackman2009-07-084-190/+83
| * | | added requeue_next_n to disk_queue and made use of it in mixed_queue:to_disk_...Matthew Sackman2009-07-083-28/+62
| * | | found a bug in the memory reports in combination with hibernation in that if ...Matthew Sackman2009-07-083-52/+78
| * | | Sorted out rabbitmqctl so that it sends pinning commands to the queue_mode_ma...Matthew Sackman2009-07-073-33/+100
| * | | lots of tuning and testing. Totally rewrote the to_disk_only_mode in mixed_qu...Matthew Sackman2009-07-076-57/+103
| * | | oh look, another merge in from 21087Matthew Sackman2009-07-061-7/+8
| |\ \ \
| * \ \ \ (another) merge from 21087 (not 21097 as mentioned in previous commit)Matthew Sackman2009-07-061-28/+49
| |\ \ \ \
| * \ \ \ \ merge from bug21097Matthew Sackman2009-07-0620-919/+3984
| |\ \ \ \ \
| | * | | | | Added documentationMatthew Sackman2009-07-061-0/+66
| | * | | | | testing shows these values work well. The whole thing works pretty well. Obv...Matthew Sackman2009-07-063-4/+4
| | * | | | | Reworked. Because the disk->mixed transition doesn't eat up any ram, there is...Matthew Sackman2009-07-038-212/+435
| | * | | | | report memory:Matthew Sackman2009-07-031-5/+5
| | * | | | | wip, dnc.Matthew Sackman2009-07-022-27/+89
| | * | | | | cosmeticMatthew Sackman2009-07-021-2/+2
| | * | | | | well if we're going to not actually pull messages off disk when going to mixe...Matthew Sackman2009-07-023-84/+29
| | * | | | | Sorted out the timer versus hibernate binary backoff. The trick is to use app...Matthew Sackman2009-07-021-10/+17
| | * | | | | merge in from 21087. Behaviour is now broken because the timeout can get > 10...Matthew Sackman2009-07-0219-908/+3675
| | |\ \ \ \ \
| | | * | | | | When converting to disk mode, use tx_publish and tx_commit instead of publish...Matthew Sackman2009-07-011-9/+16
| | | * | | | | Well after all that pain, simply doing the disk queue tests first seems to so...Matthew Sackman2009-07-011-3/+1
| | | * | | | | merge, but it still doesn't work. Sometimes it blows up on clustering with "A...Matthew Sackman2009-06-301-5/+14
| | | |\ \ \ \ \
| | | | * | | | | Well, this seems to work.bug19662Matthew Sackman2009-06-301-4/+13
| | | | * | | | | and now clustering seems to work again...Matthew Sackman2009-06-301-4/+4
View Lorry Controller Status