All Projects : Tungsten Clustering (Key: TUC)

Project Lead: Robert Hodges
URL: http://www.continuent.com/community/
Description:
JIRA list for all Tungsten clustering bugs, feature requests, and improvements

Release Notes

Road Map View personal road map Scope: next 3 versions | all versions Release 1.2.4 ( 15/Apr/10 | Release Notes ) Progress:    6 of 9 issues have been resolved Tungsten Maintenance Release    TUC-115 UNRESOLVED Router memory leak when using session consistency    TUC-123 UNRESOLVED Connector maintains client connection even though the server connection has failed/gone away    TUC-110 UNRESOLVED Router requires mysql privileges to execute 'show master status' while using session consistency    TUC-98 FIXED Session-consistency: Connector updates highwater after executing a read and updates it with the highwater from the slave    TUC-59 FIXED Severe memory leak discovered when running with standard connection pooling    TUC-92 FIXED Slave can stop replicating after applying an event that failed on the master but that was logged anyway    TUC-124 FIXED Connector must not switch to a slave, when using session consistency OR read/write splitting if a temp table has been created on a given connection.    TUC-129 FIXED In row replication, database object names should be backticked    TUC-99 FIXED Session-consistency: SQL-Router flip-flops between slaves even if the current read-channel can be used Release 1.2.5 ( Release Notes ) Progress:    0 of 6 issues have been resolved Tungsten Maintenance Release    TUC-96 UNRESOLVED Master replicator can break with "Too many open files" error    TUC-25 UNRESOLVED Replication fails to apply statements contain escape codes that confuse MySQL JDBC driver    TUC-130 UNRESOLVED In row replication, strings (varchar, text, ...) are stored in the mysql binlog using the table character set. If this character set differs from the JVM default character set, data can be corrupted at extraction time.    TUC-93 UNRESOLVED Master fails extracting a date with error message "Event extraction failed: bad date format for date value: 1029315 String: 2010-6-3"    TUC-94 UNRESOLVED MySQLStatementTranslator can be confused by a '*' character found inside a string    TUC-48 UNRESOLVED Slave is in SHUNNED state instead of recovered to ONLINE after a node reboot Release 1.3 ( 05/Mar/10 | Release Notes ) Progress:    66 of 80 issues have been resolved Flexible replication pipelines plus DBMS support and performance improvements    TUC-106 UNRESOLVED Configure script should ask for the list of managers when the new router protocol is on    TUC-119 UNRESOLVED Replication fails on very large INSERT statement    TUC-125 UNRESOLVED Replicator does not correctly fragment transactions on MySQL    TUC-128 UNRESOLVED When installing connector-only, configure puts the manager start and stop in startall and stopall, respectively, as well as listing the manager in deployall/undeployann    TUC-97 UNRESOLVED Restarting a slave after skipping some failed events can fail    TUC-51 UNRESOLVED "Lock wait timeout exceeded" error can appear when a consistency check is running against a big table    TUC-107 UNRESOLVED Add router configuration setting for URL of c3p0 pool to reset    TUC-46 UNRESOLVED Configure script does not correctly handle turning off replicator or non-standard MySQL    TUC-130 UNRESOLVED In row replication, strings (varchar, text, ...) are stored in the mysql binlog using the table character set. If this character set differs from the JVM default character set, data can be corrupted at extraction time.    TUC-111 UNRESOLVED SQL Router potentially uses two reserved connections per any connection using session consistency    TUC-24 UNRESOLVED Unable to recover replicator to a 'slave' after the node it is on is isolated from the cluster and there is no other cluster activity    TUC-36 UNRESOLVED Connector fails to send back full result of stored procedure called by prepared statement (pass through mode on)    TUC-104 UNRESOLVED Tungsten Intall Guide misses mandatory configuration values    TUC-102 UNRESOLVED Wrong sequence number displayed on a restarted slave before the first update is processed    TUC-6 FIXED Add ability to specify replication start point    TUC-85 FIXED Connectivity broken in Tunsten 1.3    TUC-91 FIXED Data consistency issues when using session consistency - write occurs on slave    TUC-7 FIXED Develop Tungsten Open Replicator framework with plugin support    TUC-20 FIXED Extract replication events from MySQLmaster through network using slave protocol    TUC-10 FIXED Implement capability to break up large transactions into fragments    TUC-19 FIXED Implement replication pipelines    TUC-56 FIXED JdbcTHLDatabase statement can be used concurrently which can cause unpredictable errors    TUC-70 FIXED NPE in THLStoreAdapter.getLastEvent when the last event is marked to be skipped    TUC-68 FIXED On slave, trep_commit_seqno table is synchronized with extractor, which is wrong    TUC-109 FIXED Router may be ONLINE with incorrect configuration    TUC-98 FIXED Session-consistency: Connector updates highwater after executing a read and updates it with the highwater from the slave    TUC-78 FIXED Slave pipeline performance problem due to contention between store and fetch operations in THL    TUC-12 FIXED Support LOAD DATA INFILE command in MySQL statement replication    TUC-39 FIXED Upgrade sequence number handling to ensure that replicator does not hang due to missing events    TUC-127 DUPLICATE Configure needs to set the manager list in router.properties if no manager is to be installed    TUC-87 FIXED Add ability to request replicator to go offline automatically following specified seqno, heartbeat, eventid, and extracted timestamp    TUC-38 FIXED Algorithm for selecting most advanced slave does not compare sequence numbers correctly    TUC-75 FIXED Batch configure (-b) fails even with valid configuration    TUC-61 FIXED Cluster heartbeat command fails if there is a shunned master    TUC-73 FIXED Filter to ignore specific tables on statement replication    TUC-82 FIXED Manager does not report replicator latency correctly when router is not running    TUC-54 FIXED NPE in TungstenPlugin.waitForAppliedEvent    TUC-2 FIXED Old masters might rejoin cluster with silently corrupt data. Potentially fix with epoch#.    TUC-88 FIXED Relay log downloading for MySQL has a hardcoded packet-size limitation of 16M    TUC-126 FIXED Replicator runs out of memory if replicator.global.buffer.size is too large    TUC-79 FIXED Rollback of 'switch' leaves the cluster non-operable    TUC-81 FIXED Running recover on live master can result in invalid datasource/replicator combination    TUC-92 FIXED Slave can stop replicating after applying an event that failed on the master but that was logged anyway    TUC-69 FIXED Switch fails intermittently due to ConnectorHandler thread dying in master replicator    TUC-77 FIXED THL event cache blocks pipeline if number of events reaches maximum cache size    TUC-23 DUPLICATE Provide a means to support application-transparent session-consistent reads    TUC-86 FIXED Add installation and configuration support for enterprise extensions    TUC-100 FIXED Connector does not properly acknowledge clients upon close commands, leading to TCP connections remaining in TIME_WAIT state    TUC-114 FIXED Disallow use of both new router gateway and old JMX protocols at the same time    TUC-95 FIXED Event fragmentation attribute for MySQL extractor needs to be expressed in bytes    TUC-47 FIXED Extract binlog event flag indicating if an event is thread specific or not.    TUC-44 FIXED Implement session variables support    TUC-129 FIXED In row replication, database object names should be backticked    TUC-60 FIXED Integration of 1.2.2 changes into the trunk    TUC-74 FIXED TUC-73 JavaScriptFilter: pass JS return value through    TUC-67 FIXED Last fragment of an event can be a Commit, which is wrongly considered as an empty commit, and therefore is not applied    TUC-49 FIXED Managers are in rules-based 'flapping' due to incorrectly detecting that they are isolated    TUC-1 FIXED Mysql extractor may skip event while displaying "Unexpected failure while extracting event" error message in log files    TUC-58 FIXED NPE in manager when running replicator status command    TUC-5 FIXED Open Replicator script plugin does not collect and print stdout for scripts    TUC-16 FIXED Parameterize commonly used variables in replicator property files    TUC-120 FIXED PostgreSQL connectivity broken in 1.3    TUC-72 FIXED Potential threading issues in manager singleton classes    TUC-4 FIXED TUC-1 Provide a setting to indicate if the extractor process should stop or skip an event in case an error occurs    TUC-40 FIXED Replicator may try to apply events which were previously skipped    TUC-99 FIXED Session-consistency: SQL-Router flip-flops between slaves even if the current read-channel can be used    TUC-53 FIXED THL events store the time when changes were extracted whereas they could store time when the events was run on the master    TUC-3 FIXED Unsigned numeric variables are not correctly replicated using row replication    TUC-13 FIXED When a rotate event is extracted, the extractor might try to open the new binlog file before it really exists on disk    TUC-71 DUPLICATE Manager does not show correct latency value    TUC-43 CANNOT REPRODUCE If no checkers are defined, the monitor fails to start and this, in turn, prevents the manager from starting.    TUC-45 NOT AN ISSUE TUC-40 Check if TUC-40 also exists into trunk    TUC-101 NOT AN ISSUE Connector should close direct connections upon database change    TUC-52 FIXED "Waiting for up to 30 seconds for the manager" ignores ^C    TUC-50 FIXED After an error while applying events, status command does not display any hints about the sequence number that brought the error    TUC-90 FIXED Extension folder not created with connector standalone install    TUC-31 FIXED Generation of full source TGZ and ZIP build    TUC-122 FIXED Include the connector in the cluster-home/bin/*all scripts    TUC-121 FIXED Remove the 'Deployment Type' configuration option    TUC-42 CANNOT REPRODUCE Manager does not load service definitions if no replicator is enabled on the node

Reports

Preset Filters

Project Summary

	Open	49	38%
	Reopened	1	1%
	Resolved	75	58%
	Closed	5	4%

Open Issues

	Blocker	10	20%
	Critical	5	10%
	Major	29	58%
	Minor	6	12%

By Assignee

Gilles Rayrat	9	18%
Robert Hodges	30	60%
Stephane Giron	10	20%
Tommi Takkunen-Lauri	1	2%