summaryrefslogtreecommitdiff
Commit message (Collapse)AuthorAgeFilesLines
* We don't need to worry about confirms, they will get sent. Change to a soft ↵Simon MacMullen2014-08-141-4/+6
| | | | error for new errors.
* Support cancel.Simon MacMullen2014-08-141-1/+9
|
* Unused var.Simon MacMullen2014-08-141-1/+1
|
* Security hole--Simon MacMullen2014-08-142-21/+25
|
* Rename this thing for consistency and also RabbitMQ-specificness.Simon MacMullen2014-08-142-6/+7
|
* Prototype of fast RPC replies.Simon MacMullen2014-08-132-5/+92
|
* stable to defaultSimon MacMullen2014-08-1354-800/+2107
|\
| * For some reason post bug 26337 we need to make sure this gets logged in the ↵Simon MacMullen2014-08-121-1/+1
| | | | | | | | server and not though rabbitmqctl.
| * Merge bug25884Simon MacMullen2014-08-121-2/+2
| |\
| | * Actually maybe that's clearer.Simon MacMullen2014-08-121-2/+2
| | |
| * | RenameSimon MacMullen2014-08-123-7/+6
| | |
| * | And max-length is implemented in terms of messages_ready, so do the same for ↵Simon MacMullen2014-08-121-1/+1
| | | | | | | | | | | | max-bytes
| * | Merge in defaultSimon MacMullen2014-08-123-31/+54
| |\ \ | | |/
| | * Merge bug26339Simon MacMullen2014-08-123-31/+54
| | |\
| | | * Document the two new terms.Simon MacMullen2014-08-121-0/+8
| | | |
| | | * Track ready and unacked bytes separately.Simon MacMullen2014-08-122-31/+46
| | |/
| * | Max-bytes argument / policy.Simon MacMullen2014-08-123-30/+47
| |/
| * Reactor - extract common fail case.Simon MacMullen2014-08-121-17/+13
| |
| * Merge bug26336Simon MacMullen2014-08-120-0/+0
| |\
| * | Improve logging when auto-clustering fails.Simon MacMullen2014-08-121-11/+23
| |/
| * Merge bug26337Simon MacMullen2014-08-127-90/+46
| |\
| | * Simplify, given the now-temporary nature of this data structure.Simon MacMullen2014-08-121-7/+3
| | |
| | * Don't log a silly warning if starting from scratch, tidy messages a bit.Simon MacMullen2014-08-111-5/+9
| | |
| | * Remove a ton of special-case error_logger invocationsSimon MacMullen2014-08-116-29/+26
| | |
| | * Remove pointless gen_server.Simon MacMullen2014-08-112-52/+11
| | |
| * | stable to defaultSimon MacMullen2014-08-1252-672/+1976
| |\ \
| | * | Explain betterSimon MacMullen2014-08-111-1/+1
| | |/
| | * Merge bug25666Simon MacMullen2014-08-119-100/+270
| | |\
| | | * OopsSimon MacMullen2014-08-111-2/+1
| | | |
| | | * Merge in defaultSimon MacMullen2014-08-119-24/+32
| | | |\ | | | |/ | | |/|
| | * | Merge bug26335Simon MacMullen2014-08-110-0/+0
| | |\ \
| | * | | Maybe a tad clearer?Simon MacMullen2014-08-111-9/+9
| | | | |
| | * | | Simplify maintenance of persistent_count on ack and purge:Simon MacMullen2014-08-111-45/+34
| | |/ / | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | * Don't parameterise remove_queue_entries/5 with a fold function, there's only one call site. * Get remove_queue_entries/4 / purge_betas_and_deltas/3 to update persistent count directly, not removed counts by store. * Update persistent count in one place in remove_pending_ack/2, not many places in ack/2. * Remove a couple of no-longer needed helpers.
| | * | stable to defaultSimon MacMullen2014-08-112-0/+10
| | |\ \
| | * \ \ stable to defaultSimon MacMullen2014-08-1149-546/+1691
| | |\ \ \
| | | * | | Dialyzer points out that the false branch cannot happen. In bug 26204 we ↵Simon MacMullen2014-08-081-4/+2
| | | | | | | | | | | | | | | | | | | | | | | | removed that distinction (cancelling an already fired timer); cancel_timer now only returns 'ok'.
| | | * | | Fix type error (introed in 24926 I think) spotted by Dialyzer.Simon MacMullen2014-08-081-2/+2
| | | | | |
| | | * | | Sigh. The type of the 4th argument to application:set_env/4 has changed in ↵Simon MacMullen2014-08-081-7/+5
| | | | | | | | | | | | | | | | | | | | | | | | Erlang 17 from being non_neg_integer() | 'infinity' to a proplist of options. And earlier versions of Erlang do not appreciate a proplist here! So just stop using it (yes, 5s timeout, but we have lots of calls to set_env/3 in the tests already and they do not seem to fail).
| | | * | | stable to defaultSimon MacMullen2014-08-081-1/+1
| | | |\ \ \
| | | * \ \ \ Merge bug26326Simon MacMullen2014-08-070-0/+0
| | | |\ \ \ \
| | | * | | | | A case missed in bug 26204.Simon MacMullen2014-08-071-2/+2
| | | |/ / / /
| | | | | | * Minor tweaks: reduce distance to default, fix spec, improve assertion, ↵Simon MacMullen2014-08-112-5/+10
| | | | | | | | | | | | | | | | | | | | | | | | | | | | improve comment, cosmetic.
| | | | | | * Add some docs for the message_bytes family.Simon MacMullen2014-08-071-0/+12
| | | | | | |
| | | | | | * Merge in defaultSimon MacMullen2014-08-078-125/+266
| | | | | | |\ | | | | |_|/ / | | | |/| | |
| | | | | | * Upgrades to the queue index and recovery terms format.Simon MacMullen2014-08-052-5/+60
| | | | | | |
| | | | | | * Fix tests, cosmetic.Simon MacMullen2014-08-052-3/+4
| | | | | | |
| | | | | | * Rethink how we do this. Previously we tried to manage bytes exactly the same ↵Simon MacMullen2014-07-291-122/+128
| | | | | | | | | | | | | | | | | | | | | | | | | | | | as len, and ram_msg_bytes the same as ram_msg_count. The trouble is that means they only account for ready messages. It would be a bit of overkill to start tracking bytes used for unacked messages separately, so instead we now only track bytes for all messages.
| | | | | | * Merge in bug26307Simon MacMullen2014-07-292-11/+21
| | | | | | |\
| | | | | | * | Also...Simon MacMullen2014-07-291-1/+1
| | | | | | | |
| | | | | | * | Merge in bug26307 and update.Simon MacMullen2014-07-2911-48/+165
| | | | | | |\ \
View Lorry Controller Status