Author: aidan Date: Tue Apr 15 05:06:43 2008 New Revision: 648216 URL: http://svn.apache.org/viewvc?rev=648216&view=rev Log: Merged revisions 627417-648207 via svnmerge from https://svn.apache.org/repos/asf/incubator/qpid/branches/M2.1
........ r627552 | rgodfrey | 2008-02-13 18:10:53 +0000 (Wed, 13 Feb 2008) | 1 line QPID-790 : Performance Improvements ........ r627789 | rupertlssmith | 2008-02-14 16:01:15 +0000 (Thu, 14 Feb 2008) | 1 line QPID-9 : Nested field tables implemented. Also wrote a test that encodes/decodes one to check it works. ........ r627794 | rgodfrey | 2008-02-14 16:24:11 +0000 (Thu, 14 Feb 2008) | 1 line QPID-790 : Performance Improvements ........ r627817 | gsim | 2008-02-14 18:02:04 +0000 (Thu, 14 Feb 2008) | 3 lines Fixed typo in output statement. ........ r627971 | gsim | 2008-02-15 08:26:00 +0000 (Fri, 15 Feb 2008) | 6 lines * updated c++ build to work with recent gentools changes * add null exchange.bound impl to SessionHandlerImpl (to reflect change to spec file) * pass AMQDataBlocks rather than AMQFrames to OutputHandler * allow client to pass messages frames to io layer in one go (via FrameList) if it will fit in a single buffer ........ r629137 | gsim | 2008-02-19 16:02:25 +0000 (Tue, 19 Feb 2008) | 3 lines Moved c++ to 0-9, based on a spec file with all the WIP stuff removed manually to avoid having to tinker with generator to exclude it. ........ r629158 | aidan | 2008-02-19 16:53:57 +0000 (Tue, 19 Feb 2008) | 2 lines Qpid-594: make AMQConnection listen for exceptions that are thrown asynchronously in it's constructor and do something appropriate with them ........ r629518 | rupertlssmith | 2008-02-20 16:04:25 +0000 (Wed, 20 Feb 2008) | 1 line QPID-800 : junit toolkit sources added. ........ r629540 | aidan | 2008-02-20 17:12:32 +0000 (Wed, 20 Feb 2008) | 2 lines QPID-786 Remove bogus Failover timeout, add test. ........ r629731 | rgodfrey | 2008-02-21 10:09:03 +0000 (Thu, 21 Feb 2008) | 1 line QPID-790 : Performance Improvements ........ r629824 | aidan | 2008-02-21 15:28:43 +0000 (Thu, 21 Feb 2008) | 4 lines QPID-785: Make sure queue browser consumers are auto-close, add test for browsing an empty queue. Refactor QueueBrowserTest a little to split up responsibilities a bit. We should move the sendMessage stuff to a super class, there are at least 4 implementations of that kicking about. ........ r629863 | rupertlssmith | 2008-02-21 16:42:27 +0000 (Thu, 21 Feb 2008) | 1 line QPID-802 : Added one minute run of benchmark tests against external broker to the build system. ........ r629981 | rgodfrey | 2008-02-21 20:42:21 +0000 (Thu, 21 Feb 2008) | 1 line QPID-804 : Fix Java Broker Python test failures ........ r630239 | rgodfrey | 2008-02-22 16:15:11 +0000 (Fri, 22 Feb 2008) | 1 line QPID-790 : Performance Improvements ........ r630247 | aidan | 2008-02-22 16:50:26 +0000 (Fri, 22 Feb 2008) | 2 lines QPID-771: fix up exception handling a bit more, this is so lovely. ........ r630723 | ritchiem | 2008-02-25 03:22:39 +0000 (Mon, 25 Feb 2008) | 1 line QPID-806 : Added startsWith and endsWith methods to AMQShortString, including test. ........ r630724 | ritchiem | 2008-02-25 03:33:58 +0000 (Mon, 25 Feb 2008) | 1 line QPID-807 : made methods public. ........ r630725 | ritchiem | 2008-02-25 03:36:55 +0000 (Mon, 25 Feb 2008) | 1 line QPID-808 : add method to qpid.jms Interface. ........ r630728 | ritchiem | 2008-02-25 03:52:54 +0000 (Mon, 25 Feb 2008) | 1 line QPID-809 : Added comments and additional error logging. ........ r630729 | ritchiem | 2008-02-25 03:58:25 +0000 (Mon, 25 Feb 2008) | 1 line QPID-107 added ACCESS_REFUSED as a reason to throw AMQAuthenticationException ........ r630730 | ritchiem | 2008-02-25 03:59:58 +0000 (Mon, 25 Feb 2008) | 1 line QPID-809 : Added comments and improved logging msgs. ........ r630731 | ritchiem | 2008-02-25 04:06:32 +0000 (Mon, 25 Feb 2008) | 1 line Adjusted the default debug level to debug (it was only info) ........ r630733 | ritchiem | 2008-02-25 04:56:42 +0000 (Mon, 25 Feb 2008) | 1 line QPID-810 : Moved check for closingChannels higher in stack and close channel on any AMQException being thrown from the body.handle methods. ........ r630833 | ritchiem | 2008-02-25 12:59:03 +0000 (Mon, 25 Feb 2008) | 1 line QPID-811 : The RejectionEE is occuring because the task pool is shutdown before the close-ok has been received. ........ r630854 | ritchiem | 2008-02-25 13:54:46 +0000 (Mon, 25 Feb 2008) | 6 lines QPID-107 : Initial ACL implementation for review. Implemented Permissions : Consume, Create, Publish. The Permissions are configured via XML in a user friendly way. Sections for consume, create and publish are currently used to further define Access and Bind internally. Access is granted to all users that have some permission. Bind rights are given to users with Create rights. Full details of the ACL design will be posted on the wiki : http://cwiki.apache.org/qpid/qpid-design-access-control-lists.html ........ r630857 | ritchiem | 2008-02-25 14:11:06 +0000 (Mon, 25 Feb 2008) | 1 line QPID-107 : Missing acl example config ........ r630865 | ritchiem | 2008-02-25 14:50:04 +0000 (Mon, 25 Feb 2008) | 1 line QPID-809 Updated logging and removed an unnecessary printStackTrace(). ........ r630872 | ritchiem | 2008-02-25 15:16:01 +0000 (Mon, 25 Feb 2008) | 1 line QPID-809 Updated logging and some whitespace changes. ........ r630873 | ritchiem | 2008-02-25 15:18:23 +0000 (Mon, 25 Feb 2008) | 1 line QPID-809 Updated logging and some whitespace changes. ........ r630892 | ritchiem | 2008-02-25 15:52:24 +0000 (Mon, 25 Feb 2008) | 1 line QPID-107 : Fixed the left over ACL'd ApplicationRegistry from killing subsequent tests. ........ r630896 | ritchiem | 2008-02-25 15:58:38 +0000 (Mon, 25 Feb 2008) | 2 lines QPID-809 Updated ignore list ........ r631141 | ritchiem | 2008-02-26 09:15:31 +0000 (Tue, 26 Feb 2008) | 1 line QPID=813 : Synchronized getInstance ........ r631264 | rgodfrey | 2008-02-26 15:55:29 +0000 (Tue, 26 Feb 2008) | 1 line QPID-804 : Fix Java Broker Python test failures ........ r631295 | rupertlssmith | 2008-02-26 16:57:05 +0000 (Tue, 26 Feb 2008) | 1 line QPID-800 : Fixed copyright notices that I had previously missed. ........ r631296 | ritchiem | 2008-02-26 16:57:29 +0000 (Tue, 26 Feb 2008) | 1 line QPID-107 : Fixed the left over old AccessManager/PrincipalDatabase references. ........ r631651 | ritchiem | 2008-02-27 17:11:45 +0000 (Wed, 27 Feb 2008) | 1 line Updated ignore list for eclipse files. ........ r631936 | rgodfrey | 2008-02-28 12:05:16 +0000 (Thu, 28 Feb 2008) | 1 line QPID-818 : Persistent Pub/Sub can get exception on acking message ........ r632054 | ritchiem | 2008-02-28 16:51:12 +0000 (Thu, 28 Feb 2008) | 1 line QPID-821 Set default to false for MultiIO and Buffer Limiting. ........ r632055 | ritchiem | 2008-02-28 16:58:43 +0000 (Thu, 28 Feb 2008) | 1 line QPID-822 Set default to false for JMSXUserID. ........ r632070 | ritchiem | 2008-02-28 17:38:20 +0000 (Thu, 28 Feb 2008) | 1 line QPID-822 Setting default to false for JMSXUserID cause this test to fail. Now test only validates the JMSXUserID if it exists. ........ r632363 | ritchiem | 2008-02-29 16:00:19 +0000 (Fri, 29 Feb 2008) | 1 line QPID-702 Removed cluster code as it has been scheduled for such since last year. ........ r633088 | ritchiem | 2008-03-03 14:02:27 +0000 (Mon, 03 Mar 2008) | 1 line QPID-594 QPID-107 Updated Connection missing AMQAuthenticationException problems ........ r633099 | aidan | 2008-03-03 14:35:58 +0000 (Mon, 03 Mar 2008) | 2 lines Import test fail script ........ r633103 | aidan | 2008-03-03 14:40:48 +0000 (Mon, 03 Mar 2008) | 2 lines Add licensce header ........ r633111 | ritchiem | 2008-03-03 14:54:59 +0000 (Mon, 03 Mar 2008) | 1 line QPID-594 QPID-107 Updated Connection missing AMQAuthenticationException problems ........ r633160 | ritchiem | 2008-03-03 16:28:36 +0000 (Mon, 03 Mar 2008) | 1 line QPID-784 : Minor changes based on code comments by ASkinner and RGodfrey. ........ r633201 | ritchiem | 2008-03-03 17:44:03 +0000 (Mon, 03 Mar 2008) | 1 line QPID-107 : Updated two test cases that can fail due to the session being closed before we can attempt to use the connection. Also there is no way to get any exception that has occurred on the connection ........ r633256 | rgodfrey | 2008-03-03 20:13:55 +0000 (Mon, 03 Mar 2008) | 1 line QPID-828 : Stop transient message data being cleared while still delivering ........ r633461 | rgodfrey | 2008-03-04 13:01:54 +0000 (Tue, 04 Mar 2008) | 1 line QPID-831 : Remove incorrect references to getDeliveredToConsumer() ........ r634697 | ritchiem | 2008-03-07 14:43:14 +0000 (Fri, 07 Mar 2008) | 1 line QPID-840 : fixed the GC setting. ........ r634717 | rgodfrey | 2008-03-07 15:35:07 +0000 (Fri, 07 Mar 2008) | 1 line QPID-839 : Fixed the hanging issue with this test; as yet unable to reproduce the test failure ........ r634720 | rgodfrey | 2008-03-07 15:54:41 +0000 (Fri, 07 Mar 2008) | 1 line QPID-839 : DUPS_OK should behave exactly as AUTO_ACK ... this would not cause the seen test failure; but is still incorrect ........ r634751 | ritchiem | 2008-03-07 17:18:25 +0000 (Fri, 07 Mar 2008) | 2 lines QPID-841 : Removed SimpleACLTest from the build until we can diagnose why it occasionally fails to find a file on disk ........ r635549 | ritchiem | 2008-03-10 13:04:21 +0000 (Mon, 10 Mar 2008) | 1 line QPID-847 : SelectorParserTest with disabled testLike() ........ r635590 | ritchiem | 2008-03-10 16:27:07 +0000 (Mon, 10 Mar 2008) | 1 line QIPD-841 : Problem with SimpleACLTest was that the ConfigurationFilePrincipalDatabaseManager was looking up the ApplicationRegistry for the configuration. Most of the time this is ok but during test tearDown/startUp sometimes the config is null on lookup. This method is just wrong anyway. the ConfigurationFilePrincipalDatabaseManager setup is the only setup that looksup the configuration from the AppRegistry. So adjusted so that it takes it as a parameter as other classes do. ........ r635602 | ritchiem | 2008-03-10 17:16:09 +0000 (Mon, 10 Mar 2008) | 1 line QPID-107 : Changes based on code review. ........ r635611 | ritchiem | 2008-03-10 17:30:14 +0000 (Mon, 10 Mar 2008) | 1 line QPID-107 : Over zealous change set reduction removed an required import ........ r636273 | ritchiem | 2008-03-12 10:40:01 +0000 (Wed, 12 Mar 2008) | 1 line QPID-845 : Reinserted ':' to JMSMessageID. ........ r636347 | ritchiem | 2008-03-12 14:44:40 +0000 (Wed, 12 Mar 2008) | 1 line QPID-846 : Update to prevent JMSX* values being masked, updated JMSPropertiesTest to prevent regression. ........ r637047 | ritchiem | 2008-03-14 10:46:40 +0000 (Fri, 14 Mar 2008) | 1 line QPID-592 : Parameterised the Read/Write buffer limits. On the broker extra config [read|write]BufferLimitSize on the client System properties qpid.[read|write].buffer.limit. All the defaults are 256k(262144). ........ r637048 | ritchiem | 2008-03-14 10:51:40 +0000 (Fri, 14 Mar 2008) | 1 line QPID-851 : Update to AMQChannel to prevent the memory leak when an autoclose consumer closes in a second thread before the register gets a chance to add the new session to the map. ........ r637066 | ritchiem | 2008-03-14 11:36:42 +0000 (Fri, 14 Mar 2008) | 1 line QPID-852 : Updated broker so that it closes consumers when there are no messages on the queue. ........ r637067 | ritchiem | 2008-03-14 11:39:20 +0000 (Fri, 14 Mar 2008) | 1 line QPID-852 : Updated broker so that it closes consumers when there are no messages on the queue ........ r637073 | ritchiem | 2008-03-14 11:54:08 +0000 (Fri, 14 Mar 2008) | 1 line QPID-852 : Missed this file in the commit as the interface had changed. ........ r637084 | ritchiem | 2008-03-14 12:45:59 +0000 (Fri, 14 Mar 2008) | 1 line QPID-853 : Use the result of processMethod (match) to decide if we have the right frame. ........ r637086 | ritchiem | 2008-03-14 12:57:42 +0000 (Fri, 14 Mar 2008) | 1 line QPID-854 : Changes to the client to make the dispatcher responsible for closing the queue browser when all the messages have been processed. ........ r637137 | aidan | 2008-03-14 15:57:40 +0000 (Fri, 14 Mar 2008) | 2 lines QPID-839 Use AMQish things to get the queue depth, elminating false negatives ........ r637146 | aidan | 2008-03-14 16:29:08 +0000 (Fri, 14 Mar 2008) | 2 lines QPID-848: Add short sleep when a connection fails to complete so that we wait for all exceptions to be thrown ........ r637170 | ritchiem | 2008-03-14 17:31:44 +0000 (Fri, 14 Mar 2008) | 1 line QPID-855 : Added check for the IllegalStateException and ignore ones from the Noop Operation. ........ r637172 | ritchiem | 2008-03-14 17:34:39 +0000 (Fri, 14 Mar 2008) | 1 line QPID-854 : Added more tests for QueueBrowser ........ r637176 | ritchiem | 2008-03-14 17:39:16 +0000 (Fri, 14 Mar 2008) | 1 line QPID-854 : This test required a change as a result of the QueueBrowser Tests. ........ r637828 | ritchiem | 2008-03-17 10:36:13 +0000 (Mon, 17 Mar 2008) | 1 line QPID-856 : Moved unbind out of the synchronized block as it doesn't need to be done there. ........ r637896 | ritchiem | 2008-03-17 14:22:13 +0000 (Mon, 17 Mar 2008) | 1 line QPID-850 : Close client connection if it is appropriate. This was hanging the tests because VMTestCase has a dummy test that causes a new connection to be init'd but never closed ........ r637977 | ritchiem | 2008-03-17 17:22:23 +0000 (Mon, 17 Mar 2008) | 1 line QPID-847 : InvalidSelectorException. This was caused by the Broker now being more AMQP spec compliant than previously. Reverted the change in AMQMinaProtocolSession.java that is causing the issue but we need to correctly fix this issue in the client as the client is not AMQP spec compliant, even with the STRICT_AMQP flag. Updated SelectorTest.java with an additional test so we don't have the functionality reversion later. ........ r637979 | ritchiem | 2008-03-17 17:25:12 +0000 (Mon, 17 Mar 2008) | 1 line QPID-854 : Fixed QueueBrowser tests that use ClientAck and Transacted, which appear to leave msgs behind per test. Even though each test should be on a new VM broker. ........ r637986 | ritchiem | 2008-03-17 17:40:14 +0000 (Mon, 17 Mar 2008) | 1 line QPID-849 : Client Deadlock, there are various points where we take the failover mutex to check the _closed values which if it is false then that is all we do. As the _closed check doesn't require the mutex then move all the checks out side of the mutex lock. ........ r637989 | ritchiem | 2008-03-17 17:44:56 +0000 (Mon, 17 Mar 2008) | 1 line QPID-857 : Reset the State Manager when receiveing a CLOSED state. ........ r637991 | ritchiem | 2008-03-17 17:46:54 +0000 (Mon, 17 Mar 2008) | 1 line QPID-858 : Added cancel to the housekeeping thread when the vhost is closed. ........ r637992 | ritchiem | 2008-03-17 17:48:24 +0000 (Mon, 17 Mar 2008) | 2 lines QPID-841,QPID-854 : Reinstated SimpleACLTest and removed QueueBrowserTest from the junit-toolkit run for now as the class has been renamed ........ r638322 | ritchiem | 2008-03-18 11:16:31 +0000 (Tue, 18 Mar 2008) | 1 line QPID-854 : Disabled one of the failover QueueBrowser tests as I believe it to be a problem with mina and inVM broker. ........ r638346 | ritchiem | 2008-03-18 12:33:34 +0000 (Tue, 18 Mar 2008) | 1 line QPID-847 : Prevented the InvalidArgumentException from closing the connection. ........ r638412 | ritchiem | 2008-03-18 15:20:15 +0000 (Tue, 18 Mar 2008) | 1 line QPID-854 : Renabled the test as the cause was the lack of closing of the purgeQueue connection in VMTestCase ........ r638760 | ritchiem | 2008-03-19 09:39:50 +0000 (Wed, 19 Mar 2008) | 1 line QPID-854 : Tests were failing as the new connection hadn't been started. Also replaced purgeQueue with checkQueuesClean as the tests should ensure they have tidied up.. if they haven't then something has gone wrong. So purging will just hide that issue. Used a second connection in the QBAAT to validate the messages to be sure they are still there. This also fixes the issues with Transacted and ClientAck where the messages would of course go back on the queue as no ack or commit were performed. Using a new connection is also cleaner as it ensures the queue browsing has no side effects for other users of the system. ........ r638799 | aidan | 2008-03-19 11:48:56 +0000 (Wed, 19 Mar 2008) | 4 lines QPID-854 AcknowledgeTest clean up messages that are left on the queue VMTestCase remove bogus DummyTestCase which causes problems with a lot of tests now that we check for queue cleanliness ........ r638805 | ritchiem | 2008-03-19 12:05:39 +0000 (Wed, 19 Mar 2008) | 1 line QPID-854 : Removed cleanQueue requirement for now as a lot of tests are not cleaning up correctly. ........ r638816 | ritchiem | 2008-03-19 12:50:53 +0000 (Wed, 19 Mar 2008) | 2 lines QPID-841 : SimpleACLTest didn't close a client connection ........ r638926 | aidan | 2008-03-19 16:30:11 +0000 (Wed, 19 Mar 2008) | 2 lines QPID 854 Add a connect delay to FailOverBaseCase so we don't end up in a race condition where mina closes the broker. ........ r638950 | ritchiem | 2008-03-19 17:42:42 +0000 (Wed, 19 Mar 2008) | 1 line QPID-862 : Add IOException to valid reason to start failover and removed the propogation of the error to client, as they don't need to know of the error unless failover fails. ........ r638951 | ritchiem | 2008-03-19 17:43:37 +0000 (Wed, 19 Mar 2008) | 2 lines QPID-847 : Fixed this test for the selectors ........ r639251 | aidan | 2008-03-20 11:29:56 +0000 (Thu, 20 Mar 2008) | 2 lines QPID-854 Resynchronise the _acceptor so that we don't hang ........ r639269 | ritchiem | 2008-03-20 12:49:13 +0000 (Thu, 20 Mar 2008) | 1 line QPID-854 : Renamed logger class to be correct, Reverting change from 637176 ........ r639598 | ritchiem | 2008-03-21 11:12:29 +0000 (Fri, 21 Mar 2008) | 1 line QPID-866 : Based on Patch from ASkinner. Only the FailoverException makes sence to process this way so remove list and synchronized so we either do an add or throw the set FailoverException. ........ r639619 | aidan | 2008-03-21 13:04:46 +0000 (Fri, 21 Mar 2008) | 2 lines QPID-867 Always close connections ........ r639629 | ritchiem | 2008-03-21 13:17:11 +0000 (Fri, 21 Mar 2008) | 1 line QPID-868 : Removed the acl.config.xml from systests/etc and updated the systests pom file to locate the one in broker/etc. ........ r639684 | ritchiem | 2008-03-21 15:53:53 +0000 (Fri, 21 Mar 2008) | 1 line QPID-868 : The acl.config.xml was using Base64MD5 PF which cannot be used inVM as the SASL manager complains about the registration of the client and broker moudules. They appear to overwrite each other. ........ r640417 | rgodfrey | 2008-03-24 13:49:06 +0000 (Mon, 24 Mar 2008) | 1 line QPID-873 : Authentication Exception should be hard error; also NPE in PropertiesPrincipalDatabase when user not known ........ r640422 | ritchiem | 2008-03-24 14:15:39 +0000 (Mon, 24 Mar 2008) | 1 line QPID-872 : Adjusted TimeToLive test receive calls to use a wait of 1s(1000) as 100ms was unreasonably short. This was causing test failures on a slower machine where no failure had really occurred. ........ r640426 | aidan | 2008-03-24 14:36:36 +0000 (Mon, 24 Mar 2008) | 2 lines QPID-874 use getQueueDepth properly ........ r640434 | ritchiem | 2008-03-24 15:03:00 +0000 (Mon, 24 Mar 2008) | 1 line QPID-875 : Dropped logging level to debug. ........ r640601 | aidan | 2008-03-24 22:16:30 +0000 (Mon, 24 Mar 2008) | 1 line QPID-876 remove .exe for hot cross version compatability action ........ r643032 | ritchiem | 2008-03-31 16:41:39 +0100 (Mon, 31 Mar 2008) | 2 lines QPID-890 : Removed old references to VHostPrincipalDatabase and an errant old.PrincipalDatabaseAccessManager change. This file should now work. ........ r643153 | aidan | 2008-03-31 22:17:57 +0100 (Mon, 31 Mar 2008) | 1 line Add licensces ........ Modified: incubator/qpid/branches/thegreatmerge/ (props changed) Propchange: incubator/qpid/branches/thegreatmerge/ ------------------------------------------------------------------------------ --- svnmerge-integrated (original) +++ svnmerge-integrated Tue Apr 15 05:06:43 2008 @@ -1 +1 @@ -/incubator/qpid/branches/M2.1:1-627416 /incubator/qpid/trunk:1-647797 +/incubator/qpid/branches/M2.1:1-648207 /incubator/qpid/trunk:1-647797
