Hi, It looks like when using a single controller configuration, the virtual database config file should not contain a <Distribution> element. When we removed this from our config file, the warnings regarding the total order queue went away.
However, we are still see the other set of warnings: 2007-04-06 16:05:54,176 WARN backend.DatabaseBackend.node2 No transaction medatada found for transaction 891 releasing locks manually Any ideas as to why this warning is showing up? Thanks, Joe On 4/5/07, Joe Chen <[EMAIL PROTECTED]> wrote:
Hi Emmanuel, Please let me know if you see any obvious errors in our configuration. Here's our controller.xml: <?xml version="1.0" encoding="UTF-8" ?> <!DOCTYPE SEQUOIA-CONTROLLER PUBLIC "-//Continuent//DTD SEQUOIA-CONTROLLER 3.0-beta2 //EN" " http://sequoia.continuent.org/dtds/sequoia-controller-3.0-beta2.dtd";> <SEQUOIA-CONTROLLER> <Controller name="c1" jdbcIpAddress="10.100.50.50" jdbcPort="25322"> <JmxSettings jmxIpAddress="10.100.50.50 " jmxPort="1090"/> <Report/> </Controller> </SEQUOIA-CONTROLLER> Here's our virtual database xml: <?xml version="1.0" encoding="UTF-8"?> <!DOCTYPE SEQUOIA PUBLIC "-//Continuent//DTD SEQUOIA 3.0-beta2//EN" " http://sequoia. continuent.org/dtds/sequoia-3.0-beta2.dtd "> <SEQUOIA> <VirtualDatabase name="mydb"> <Distribution> <MessageTimeouts/> </Distribution> <Backup> <Backuper backuperName="mysql" className=" org.continuent.sequoia.controller.backup.backupers.MySQLBackuper"/> </Backup> <AuthenticationManager> <AdminUser username="admin" password=""/> <VirtualUser vLogin="user" vPassword=""/> </AuthenticationManager> <DatabaseBackend name="node1" driver="com.mysql.jdbc.Driver" url="jdbc:mysql://10.100.50.51:3306/db1" connectionTestStatement="select 1"> <ConnectionManager vLogin="user" rLogin="user" rPassword=""> <VariablePoolConnectionManager initPoolSize="20" minPoolSize="10" maxPoolSize="100" idleTimeout="30" waitTimeout="10"/> </ConnectionManager> </DatabaseBackend> <DatabaseBackend name="node2" driver="com.mysql.jdbc.Driver" url="jdbc:mysql://10.100.50.52:3306/db2" connectionTestStatement="select 1"> <ConnectionManager vLogin="user" rLogin="user" rPassword=""> <VariablePoolConnectionManager initPoolSize="20" minPoolSize="10" maxPoolSize="100" idleTimeout="30" waitTimeout="10"/> </ConnectionManager> </DatabaseBackend> <DatabaseSchema/> <RequestManager> <RequestScheduler> <RAIDb-1Scheduler level="passThrough"/> </RequestScheduler> <RequestCache> <MetadataCache/> <ParsingCache/> </RequestCache> <LoadBalancer> <RAIDb-1> <WaitForCompletion policy="first"/> <MacroHandling> <RandomMacro macroName="rand()"/> <DateMacro macroName="now()" dateFormat="timestamp"/> <DateMacro macroName="current_date()" dateFormat="date"/> <DateMacro macroName="current_date" dateFormat="date"/> <DateMacro macroName="timeofday()" dateFormat="time"/> <DateMacro macroName="current_timestamp()" dateFormat="timestamp"/> <DateMacro macroName="current_timestamp" dateFormat="timestamp"/> <DateMacro macroName="current_time()" dateFormat="time"/> <DateMacro macroName="current_time" dateFormat="time"/> </MacroHandling> <RAIDb-1-LeastPendingRequestsFirst/> </RAIDb-1> </LoadBalancer> <EmbeddedRecoveryLog/> </RequestManager> </VirtualDatabase> </SEQUOIA> As for the sample query, there's nothing really unique about the query we were doing. In one example, the query is simply selecting some address information from the db. 2007-03-30 18:46:06,110 ERROR sequoia.controller.loadbalancer Request was not found in total order queue, posting out of order ( select address0_.id as id0_0_, address0_.address as address0_0_, address0_.country as country0_0_, address0_.city as city0_0_, address0_.postalCode as postalCode0_0_, address0_.province as province0_0_ from Address address0_ where address0_.id=? /<!%L|110|!%>) Thanks, Joe On 4/3/07, Emmanuel Cecchet <[EMAIL PROTECTED]> wrote: > > Hi Joe, > > > We have Sequoia 3.0 beta 2 set up in a RAIDb1 configuration running > > with one controller and two MySQL database backends. We are also > > using Hibernate 3.2 in our web application. As far as I can tell, > > database changes are being properly replicated to both backends. > > However, during runtime, we are seeing a LOT of warnings repeatedly > > showing up in the logs. > > > > Anyone know why these warnings are showing up? Are the following > > warnings harmless or do they indicate that there is a problem with the > > setup? > It is strange that the request shows as a distributed request in a > single controller scenario. This is probably harmless but not nice, we > need to fix this. Adding a Distribution element would probably solve the > issue but add some overhead. > Could you post your configuration file and a sample query that would > generate the problem? > > Many thanks for your feedback, > Emmanuel > > -- > Emmanuel Cecchet > Chief Scientific Officer, Continuent > > Blog: http://emanux.blogspot.com/ > Open source: http://www.continuent.org > Corporate: http://www.continuent.com > Skype: emmanuel_cecchet > Cell: +33 687 342 685 > > > _______________________________________________ > Sequoia mailing list > [email protected] > https://forge.continuent.org/mailman/listinfo/sequoia >
_______________________________________________ Sequoia mailing list [email protected] https://forge.continuent.org/mailman/listinfo/sequoia
