summaryrefslogtreecommitdiff
Commit message (Collapse)AuthorAgeFilesLines
* add upgrade stepMatthias Radestock2011-12-011-0/+8
|
* tie up some loose ends for the new tableMatthias Radestock2011-11-301-5/+5
|
* now that we have table locks there is no need for phase separationMatthias Radestock2011-11-301-66/+32
| | | | | This is essentially reverting to the code as it was before bug 23885, except using the new rabbit_topic_trie_node table for counts.
* only lock tables when we are dealing with multiple bindingsMatthias Radestock2011-11-301-6/+11
|
* simplifyMatthias Radestock2011-11-301-18/+6
|
* reduce time complexity of binding removal on queue deletionMatthias Radestock2011-11-303-19/+68
| | | | | | | | | | | | from O(n^2) to O(n). The solution comes in two parts: 1) a new table, rabbit_topic_trie_node, to explicitly track edge and binding counts of nodes in the trie. This eliminates expensive mnesia:match_object operations to determine these counts when needed. 2) table-scope write locks to eliminate row-level locking, which is expensive when there are many such locks in a single transaction
* comment out failing test to check whether others fail tooMatthias Radestock2011-11-291-1/+3
|
* Merged bug24578 into defaultEmile Joubert2011-11-291-1/+1
|\
| * Don't add message id to the set of msgs we've heard of from gm but not ↵Simon MacMullen2011-11-221-1/+1
| | | | | | | | channel, if we actually *have* heard of it from the channel.
* | Merge bug 24392Simon MacMullen2011-11-291-7/+12
|\ \
| * \ Merge bug24579Simon MacMullen2011-11-294-5/+50
| |\ \
| * | | Slight simplification that these vars are declared once and failure is a thunk.Simon MacMullen2011-11-291-5/+7
| | | |
| * | | RearrangeEmile Joubert2011-11-281-18/+15
| | | |
| * | | Allow equivalence comparisons with absent valuesEmile Joubert2011-11-221-12/+18
| | |/ | |/|
* | | ensure that rabbit dies when recovery failsMatthias Radestock2011-11-291-1/+4
| | | | | | | | | | | | rather than just sitting there
* | | ignore empty segment filesMatthias Radestock2011-11-281-11/+14
| |/ |/| | | | | We already ignore corrupted content.
* | Merged bug24593 into defaultEmile Joubert2011-11-280-0/+0
|\ \
* | | Still nicer error handling inspired by lib:eval_str/1.Simon MacMullen2011-11-251-4/+10
| | |
* | | Also handle tokenising errors somewhat.Simon MacMullen2011-11-241-8/+13
| | |
* | | Handle RPCs that crash better.Simon MacMullen2011-11-241-1/+1
| | |
* | | Handle the most common parsing error nicely.Simon MacMullen2011-11-241-4/+8
| | |
* | | Basic rabbitmqctl eval, with let-it-crash error handling.Simon MacMullen2011-11-242-0/+23
|/ /
* | better reporting of location of failed testMatthias Radestock2011-11-241-2/+2
| |
* | Another attempt at reducing racy failures.Simon MacMullen2011-11-241-1/+1
| |
* | Merge bug 24467 (Autogenerated queue names don't pass validation)Simon MacMullen2011-11-231-1/+8
|\ \ | |/ |/|
| * base64url-encode for string guid generationMatthias Radestock2011-11-231-1/+8
|/ | | | | this works well for AMQP (with its restrictions on queue name characters) but also when used as part of filenames, urls, etc
* cosmeticMatthias Radestock2011-11-211-5/+6
|
* Merged bug23052 into defaultEmile Joubert2011-11-175-19/+22
|\
| * Merged bug24559 into defaultEmile Joubert2011-11-161-13/+7
| |\
| | * Put the real changes in :-(Steve Powell2011-11-151-8/+6
| | |
| | * Merge default into bug24559Steve Powell2011-11-152-6/+9
| | |\
| | * | Portfile updates for macports packagingSteve Powell2011-11-151-13/+9
| | | |
| * | | don't explain why we don't do something that doesn't workMatthias Radestock2011-11-151-4/+1
| | |/ | |/|
| * | Merged bug24560 into defaultEmile Joubert2011-11-151-1/+1
| |\ \
| | * | DuhSimon MacMullen2011-11-151-2/+1
| | | |
| | * | If we get an explicit channel close we might already be shutting the ↵Simon MacMullen2011-11-151-0/+1
| | |/ | | | | | | | | | connection down - in which case we should check if we have no more channels left and close_ok the connection if so.
| * | Merged bug23462 into defaultEmile Joubert2011-11-150-0/+0
| |\ \ | | |/
| * | More clarity.Simon MacMullen2011-11-151-2/+3
| | |
| * | Use {node(), make_ref()} rather than self().Simon MacMullen2011-11-151-7/+9
| |/
| * Added tag rabbitmq_v2_7_0 for changeset 353c527677e1Steve Powell2011-11-090-0/+0
| |
| * Updated changelogs by release buildrabbitmq_v2_7_0Steve Powell2011-11-092-0/+9
| |
* | No longer download a garbled copy of the install / build instructions from ↵Simon MacMullen2011-11-095-27/+4
|/ | | | the website.
* Merged bug24534 into defaultEmile Joubert2011-11-071-7/+65
|\
| * Merge bug24430.Simon MacMullen2011-11-0732-1197/+1011
| |\
| * | CosmeticSimon MacMullen2011-11-071-8/+6
| | |
| * | send_after does not accept infinity as an argumentEmile Joubert2011-11-031-14/+22
| | |
| * | Correct thinko - permanent + normal is an error, duh!Simon MacMullen2011-11-021-5/+6
| | | | | | | | | | | | (And some slightly evil formatting, but I really don't want this to wrap.)
| * | OopsSimon MacMullen2011-11-021-1/+1
| | |
| * | Refactor: reduce distance from default.Simon MacMullen2011-11-021-7/+4
| | |
| * | Refactor: don't use a dict when a list will do.Simon MacMullen2011-11-021-5/+7
| | |
View Lorry Controller Status