Commit Graph

83 Commits

Author SHA1 Message Date
gtully 140ce1bc8f https://issues.apache.org/jira/browse/AMQ-4485 - reenable test. concurrentStoreandDispatch case cannot be reconciled via setBatch, best we can do is trap duplicates from inflight messages as they occur. tagged async stores to support this. the revert of serialization with tx and non aysnc send brings perf back to where it was. https://issues.apache.org/jira/browse/AMQ-5266 2014-10-07 14:50:41 +01:00
gtully a0c42a61dd https://issues.apache.org/jira/browse/AMQ-5354 - patch applied with thanks 2014-09-15 15:59:00 +01:00
gtully 8cdb5c2c1d https://issues.apache.org/jira/browse/AMQ-5266 - remove err message print on iterator limit 2014-09-11 17:07:35 +01:00
gtully 54e2e3bef2 rework cursor store sync w.r.t to index order. resolve issues with skipped dispatch and duplicate dispatch. https://issues.apache.org/jira/browse/AMQ-4485 https://issues.apache.org/jira/browse/AMQ-5266 2014-08-30 00:51:21 +01:00
Timothy Bish 582fc08019 https://issues.apache.org/jira/browse/AMQ-5072
Add accessors to the persitence adapter class and a test to validate.
2014-08-14 11:14:59 -04:00
Daniel Kulp 06bb7a2716 Update poms to fully load into M2E 2014-07-10 13:47:55 -04:00
Timothy Bish 74846bb2b4 https://issues.apache.org/jira/browse/AMQ-3758
Refactor the scheduler store into a more KahaDB style store that can
recover from various problems like missing journal files or corruption
as well as rebuild its index when needed.  Move the scheduler store into
a more configurable style that allows for users to plug in their own
implementations.  Store update from legacy versions is automatic.
2014-07-07 12:28:11 -04:00
gtully 27b3a7c344 https://issues.apache.org/jira/browse/AMQ-5212 - ensure uncontented access to concurrent destination map avoides deadlock, rerework https://issues.apache.org/jira/browse/AMQ-4952 to differenciate duplicates from send and recovered messages from the store. https://issues.apache.org/jira/browse/AMQ-3454 benefits from getDestinationMap(destination) to get direct access to the map to determine existance. Additional test 2014-06-09 12:32:33 +01:00
Dejan Bosanac fa086396a7 [maven-release-plugin] prepare for next development iteration 2014-06-05 11:59:29 +02:00
Dejan Bosanac 58dd93e42e [maven-release-plugin] prepare release activemq-5.10.0 2014-06-05 11:59:17 +02:00
Dejan Bosanac c2cf78542f [maven-release-plugin] prepare for next development iteration 2014-06-04 11:33:17 +02:00
Dejan Bosanac 5bc741f227 [maven-release-plugin] prepare release activemq-5.10.0 2014-06-04 11:33:03 +02:00
Timothy Bish bfb1778211 https://issues.apache.org/jira/browse/AMQ-5166
Apply fix for tracker settings.
2014-05-06 16:31:27 -04:00
gtully ad1f751a41 https://issues.apache.org/jira/browse/AMQ-5077 - reduce reader thread work when client uses async send; async store updates can now queue up to the destination memory limit b/c they don't block the send thread. Pending store writes are now tracked in memory usage. This allows a client to quickly provide a burst of messages to fill the destination cache bounded only by network bandwidth 2014-04-30 16:13:18 +01:00
Hadrian Zbarcea 3f32507f78 Use autocrlf consistently for line endings 2014-04-29 14:18:37 -04:00
gtully 79ba2a79c1 have kahadb journal transaction commands respect enableJournalDiskSyncs - so consistent with leveldbstore - syncs off breaks the jms durability guarantee across the board - no good reason to distinguish transactional from non transaction ops; the server qos can override the client 2014-03-26 16:37:23 +00:00
gtully 266d23ef79 revisit https://issues.apache.org/jira/browse/AMQ-3519 with more durable solution https://issues.apache.org/jira/browse/AMQ-5068 - JMSRedelivered header now persisted before dispatch so that it is a reliable indication of a possible duplicate delivery. The option is enabled via destination policy persistJMSRedelivered 2014-03-26 11:16:42 +00:00
gtully cfe099d1cc rework https://issues.apache.org/jira/browse/AMQ-3305 in the context of https://issues.apache.org/jira/browse/AMQ-4952 dlq processing of duplicates, also possible missed dispatch with interleaved xa completion and new messages - recovered transactions are now individually tracked by the store and replayed in the next batch, which negates the need to flush the cursor and avoids duplicates 2014-03-05 14:51:05 +00:00
gtully ab01ae3818 fix build revert extras in MessageDatabase from https://git-wip-us.apache.org/repos/asf?p=activemq.git;a=commit;h=69c0d399 - sorry 2014-03-03 16:20:21 +00:00
gtully 69c0d399fb make perpared xa transactions visible in kahadb persistenceadapter view mbean 2014-03-03 13:55:27 +00:00
Hiram Chirino 232b8c51e9 Implements AMQ-5072: Support configuring a different directory for the KahaDB index files. 2014-02-21 11:08:27 -05:00
Hadrian Zbarcea 7b555c2898 AMQ-5047. Fix packaging issues related to maven version 2014-02-15 16:15:04 -05:00
Kevin Earls c027552035 Upgraded surefire plugin to 2.16. AMQ-5006 2014-01-30 14:22:44 +01:00
Timothy Bish dd0c9c4144 https://issues.apache.org/jira/browse/AMQ-4988 2014-01-22 10:09:38 -05:00
Timothy Bish 11ae61f539 https://issues.apache.org/jira/browse/AMQ-4969 2014-01-14 18:48:40 -05:00
Timothy Bish ef619b6a9b https://issues.apache.org/jira/browse/AMQ-4947
Updated fix for this issue allows for enabling non-forced metadata
updates to the file channel via FileChannel#force(false)

enable this by defining
"org.apache.activemq.kahaDB.files.skipMetadataUpdate=true"
2014-01-09 17:21:56 -05:00
gtully f92d45bec1 https://issues.apache.org/jira/browse/AMQ-4952 - deal with duplicates by redirecting to dlq when they are detected by the cursors such that they don't linger for redispatch after a restart. Networks are the main culprit for such duplicates b/c the producer audit traps regular failover resends 2014-01-07 16:52:39 +00:00
Timothy Bish 97b12c7245 Revert "Fix for: https://issues.apache.org/jira/browse/AMQ-4947"
This reverts commit c50b6c39ba.
2014-01-06 09:32:46 -05:00
Timothy Bish 7db0fe6a38 https://issues.apache.org/jira/browse/AMQ-4819 2013-12-20 17:29:31 -05:00
Timothy Bish 44b25cb7a1 https://issues.apache.org/jira/browse/AMQ-4819
Slightly modified version of the patch since it was not against latest
code.  Trimmed down the changes by breaking out the logging checks into
a separate method.
2013-12-20 15:56:24 -05:00
Timothy Bish c50b6c39ba Fix for: https://issues.apache.org/jira/browse/AMQ-4947 2013-12-19 18:36:05 -05:00
gtully 06f24e2e0b https://issues.apache.org/jira/browse/AMQ-4907 - sanity check on the index when checkForCorruptJournalFiles - test and check that validates the orderindex makes sense 2013-11-28 21:14:24 +00:00
Claus Ibsen c0090f6f0c AMQ-4885: Renamed some API methods in MessageStore due typos. 2013-11-22 11:21:42 +01:00
gtully ff593bf3dd add debug logging for mkaha destination to persistence adapter mapping 2013-11-15 11:29:15 +00:00
Dejan Bosanac 5dacae368b https://issues.apache.org/jira/browse/AMQ-3725 - close delegated file only if not null 2013-11-01 14:03:20 +01:00
Dejan Bosanac 582af3e74a https://issues.apache.org/jira/browse/AMQ-3725 - allow kahadb to recover from failed file system 2013-10-31 17:58:52 +01:00
gtully 864aa6798f [maven-release-plugin] prepare for next development iteration 2013-10-14 23:38:35 +01:00
gtully 87c9bbebac [maven-release-plugin] prepare release activemq-5.9.0 2013-10-14 23:38:22 +01:00
gtully ee0933c71e [maven-release-plugin] prepare for next development iteration 2013-10-11 14:51:36 +01:00
gtully b276006911 [maven-release-plugin] prepare release activemq-5.9.0 2013-10-11 14:51:24 +01:00
Hiram Chirino 28163a4065 Don't assume all persistence store use the same default directory. 2013-09-27 05:10:30 -04:00
Hiram Chirino 70c698f988 Fixes NPE that was occurring. 2013-09-27 04:51:11 -04:00
Hiram Chirino 21fe8cac7d Make the mKahaDB store agnostic to the nested persistence adapter type. 2013-09-27 04:26:17 -04:00
gtully 0f90695db7 https://issues.apache.org/jira/browse/AMQ-4729 - add lockableSupport to mKahaDb to isolate it from dynamically created filtered kahadb locks, nested locks are disabled when top level lock is in place. default to a sharedfile locker 2013-09-19 21:58:43 +01:00
Timothy Bish d1446c3bca fix and tests for: https://issues.apache.org/jira/browse/AMQ-4073 2013-09-05 15:59:01 -04:00
Timothy A. Bish 2d861dae65 fix for: https://issues.apache.org/jira/browse/AMQ-4683
Make scheduler job dispatching start more deterministic

git-svn-id: https://svn.apache.org/repos/asf/activemq/trunk@1516912 13f79535-47bb-0310-9956-ffa450edef68
2013-08-23 16:01:12 +00:00
Timothy A. Bish 85bb229aa7 fix for: https://issues.apache.org/jira/browse/AMQ-4683
git-svn-id: https://svn.apache.org/repos/asf/activemq/trunk@1516561 13f79535-47bb-0310-9956-ffa450edef68
2013-08-22 19:45:22 +00:00
Gary Tully 90d6c2042d https://issues.apache.org/jira/browse/AMQ-4212 - fix auto upgrade from ver 1 and 2 to 5 - regression in KahaDBVersionTest
git-svn-id: https://svn.apache.org/repos/asf/activemq/trunk@1511711 13f79535-47bb-0310-9956-ffa450edef68
2013-08-08 11:51:29 +00:00
Gary Tully 0061f6f755 https://issues.apache.org/jira/browse/AMQ-4000 - tracking subscription commands causes a problem with replay outside of activate - can reset a sub ack locations - regression in AMQ2149Test. Fixed by ignoring sub add during recovery replay
git-svn-id: https://svn.apache.org/repos/asf/activemq/trunk@1511306 13f79535-47bb-0310-9956-ffa450edef68
2013-08-07 13:38:53 +00:00
Gary Tully 5cadb04ad3 https://issues.apache.org/jira/browse/AMQ-4662 - kahadb default checksumJournalFiles = true - validate no migration impact as we currently store default checksum value of 0
git-svn-id: https://svn.apache.org/repos/asf/activemq/trunk@1509178 13f79535-47bb-0310-9956-ffa450edef68
2013-08-01 11:44:30 +00:00