summaryrefslogtreecommitdiff
Commit message (Collapse)AuthorAgeFilesLines
* Various fixes and improvements to client readingMatthew Sackman2010-05-051-37/+45
|
* cosmeticMatthias Radestock2010-05-051-55/+49
|
* refactor concurrent ets manipulation logicMatthias Radestock2010-05-051-43/+33
|
* made shutdown term structure opaque to qiMatthias Radestock2010-05-053-44/+40
| | | | only the vq needs to know about it
* disentangle qi from msg_storeMatthias Radestock2010-05-042-14/+10
|
* cosmeticMatthias Radestock2010-05-041-3/+2
|
* cosmeticMatthias Radestock2010-05-041-22/+18
|
* oopsMatthias Radestock2010-05-041-8/+7
|
* cosmeticMatthias Radestock2010-05-046-85/+92
|
* cosmeticMatthias Radestock2010-05-045-117/+136
|
* merge default into bug21673Matthias Radestock2010-05-040-0/+0
|\
| * Merging bug 22616 into defaultMatthew Sackman2010-05-0413-298/+2890
| |\
| | * pluggable queue backendsMatthias Radestock2010-05-036-291/+772
| | | | | | | | | | | | | | | | | | | | | | | | | | | - behaviour and specs for pluggable queue backends - reworking of the queue code to support pluggability, including hookups to memory monitoring and file handle management - interface to existing persister via the new API All these changes were cherry-picked from the bug21673 branch.
* | | Improvement to docs on BQ:syncMatthew Sackman2010-05-031-1/+3
| | |
* | | cosmeticMatthias Radestock2010-05-021-6/+6
| | |
* | | cancel queue's timers on shutdownMatthias Radestock2010-05-021-14/+20
| | | | | | | | | | | | This is cleaner, though not strictly necessary.
* | | cosmeticMatthias Radestock2010-05-023-23/+12
| | |
* | | restructure backing queue sync'ingMatthias Radestock2010-05-025-29/+31
| | | | | | | | | | | | | | | | | | | | | | | | | | | The code very much relied on the fact that sync_callback would always return the same fun when a sync was required. So it makes sense to capture that in the API by splitting sync_callback into a 'needs_sync' predicate and a separate callback handler. As a result we do not need the backing_queue_timeout_fun state member.
* | | correct backing_queue:tx_{commit,rollback} descriptionsMatthias Radestock2010-05-021-6/+4
| | |
* | | refactoring of sync timer settingMatthias Radestock2010-05-021-21/+14
| | |
* | | cosmeticMatthias Radestock2010-04-301-6/+6
| | |
* | | merge bug22616 into bug21673Matthias Radestock2010-04-300-0/+0
|\ \ \ | | |/ | |/|
| * | hook queue processes up to file handle cacheMatthias Radestock2010-04-302-1/+12
| | | | | | | | | | | | | | | | | | | | | This doesn't do anything here, since in their current form the queue processes do not use the fhc, but that will change in the future. This change was cherry-picked from the bug21673 branch.
* | | move queue's fhc registration from startup into init handlerMatthias Radestock2010-04-301-2/+2
| | | | | | | | | | | | | | | so we don't do any unnecessary work in case the queue process gets terminated before initialisation is completed
* | | cosmeticMatthias Radestock2010-04-301-9/+10
| | |
* | | Merging bug22616 into bug21673Matthew Sackman2010-04-301-8/+7
|\ \ \ | |/ /
| * | Merging default into bug22616Matthew Sackman2010-04-301-8/+7
| |\ \ | | |/
| | * Pulling good patches off bad head of defaultMatthew Sackman2010-04-301-7/+9
| | |
* | | merge bug22616 into bug21673Matthias Radestock2010-04-301-0/+2
|\ \ \ | |/ /
| * | cherry-pick file_handle_cache from bug21673Matthias Radestock2010-04-304-1/+885
| | | | | | | | | | | | and its integration with the tcp_acceptor and worker_pool code
* | | Make the msg_store_gc register an fhc callbackMatthew Sackman2010-04-301-3/+10
| | |
* | | merge bug22616 into bug21673Matthias Radestock2010-04-307-18/+37
|\ \ \ | |/ /
| * | merge default into bug22616Matthias Radestock2010-04-307-18/+37
| |\ \ | | |/
| | * bug22637 merged into default (erlang version check on start of the application)Marek Majkowski2010-04-305-15/+32
| | |\
| | | * Modify version_compare so that eg 5.0 is eq 5. On application start, check ↵Matthew Sackman2010-04-293-12/+29
| | | | | | | | | | | | | | | | that we have a sufficient version of erts
| | | * Force the RPMS to only be installable with erlang 12b3 or later (though ↵Matthew Sackman2010-04-291-2/+2
| | | | | | | | | | | | | | | | interestingly, it looks like there was no Fedora R12B-3 package - the closest I can find is R12B-4 which was in Fedora 8)
| | | * Force the debs to only be installable with erlang 12b3 or laterMatthew Sackman2010-04-291-1/+1
| | | |
| | * | bug22264 merged into default (print erts version on startup)Marek Majkowski2010-04-291-2/+3
| | |\ \ | | | |/ | | |/|
| | | * Apparently, we're not allowed to clutter the bunny. Hence added it to the ↵Matthew Sackman2010-04-291-7/+6
| | | | | | | | | | | | | | | | settings listing below
| | | * Display the erts version in the bottom right of the Rabbit bannerMatthew Sackman2010-04-291-3/+5
| | |/
| | * bug 22315 merged into default (+P 1M, process count set to 1M by default)Marek Majkowski2010-04-292-1/+2
| | |\
| | | * Bump the default number of processes to 1MMatthew Sackman2010-04-292-1/+2
| | |/
* | | Add missing specsMatthew Sackman2010-04-301-0/+2
| | |
* | | Make the logic on result of soft_close match that on hard_close (i.e. full ↵Matthew Sackman2010-04-301-43/+33
| | | | | | | | | | | | put if there's an error. Don't touch the gb_tree if there's no error. Also, revert back to two diffs for client callbacks and mrefs
* | | gb_trees:delete crashes if the key is not in the tree. I kid you not.Matthew Sackman2010-04-301-2/+2
| | |
* | | reduce_timer_set => timer_refMatthew Sackman2010-04-301-10/+10
| | |
* | | Don't allow lots of timers to be startedMatthew Sackman2010-04-302-7/+14
| | |
* | | Merging headsMatthew Sackman2010-04-301-2/+2
|\ \ \
| * | | cosmeticMatthias Radestock2010-04-301-2/+2
| | | |
* | | | Change register_callback to a cast, don't send messages in absence of a ↵Matthew Sackman2010-04-301-46/+52
|/ / / | | | | | | | | | callback, and combine two dicts
View Lorry Controller Status