Hi Emmanuel, Thanks very much for your detailed reply.
The problem is we have two controllers, and the configuration is: controller [EMAIL PROTECTED] has its own postgresql backend and the url is "jdbc:postgresql://192.168.5.1:5432/test" controller [EMAIL PROTECTED] has its own postgresql backend and the url is "jdbc:postgresql://192.168.5.1:5432/test" As you've pointed out that if I shutdown the backend posgresql @192.168.5.1, the requests are still being forwarded to controller [EMAIL PROTECTED], unless I shutdown controleer [EMAIL PROTECTED], am I right? If that's the case, I believe that's problem, initially I thought all the requrests will be forwarded to controller two automatically once I disable controller one's backend. i.e., I have to dupliate the test [EMAIL PROTECTED] and rename it in odered to be connected to controller one as well? PS: here are the configuration files I used. ------------ Controller.xml ----------- <?xml version="1.0" encoding="UTF-8" ?> <!DOCTYPE SEQUOIA-CONTROLLER PUBLIC "-//Continuent//DTD SEQUOIA-CONTROLLER 2.10.10//EN" " http://sequoia.continuent.org/dtds/sequoia-controller-2.10.10.dtd";> <SEQUOIA-CONTROLLER> <Controller ipAddress="192.168.5.2" port="25322"> <JmxSettings> <RmiJmxAdaptor port="1090"/> </JmxSettings> <VirtualDatabase configFile="postgresl-cluster.xml" virtualDatabaseName="test" autoEnableBackends="true"/> </Controller> </SEQUOIA-CONTROLLER> ------------------- postgresl-cluster.xml ----------------- <?xml version="1.0" encoding="UTF-8"?> <!DOCTYPE SEQUOIA PUBLIC "-//Continuent//DTD SEQUOIA 2.10.10//EN" " http://sequoia.continuent.org/dtds/sequoia-2.10.10.dtd";> <SEQUOIA> <VirtualDatabase name="test"> <Distribution hederaPropertiesFile="/hedera_appia.properties"> <MessageTimeouts/> </Distribution> <Backup> <Backuper backuperName="postgres" className="org.continuent.sequoia.controller.backup.backupers.PostgreSQLSplitPlainTextBackuper" options="dumpServer=192.168.5.1,encoding=UTF8,splitSize=500m"/> </Backup> <AuthenticationManager> <Admin> <User username="admin" password=""/> </Admin> <VirtualUsers> <VirtualLogin vLogin=" " vPassword=" "/> </VirtualUsers> </AuthenticationManager> <DatabaseBackend name="postgresql-prod0-test" driver="org.postgresql.Driver" url="jdbc:postgresql:// 192.168.5.1:5432/jmangotest" connectionTestStatement="select now()"> <DatabaseSchema dynamicPrecision="table"/> <ConnectionManager vLogin=" " rLogin=" " rPassword=" "> <VariablePoolConnectionManager initPoolSize="20" minPoolSize="5" maxPoolSize="0" idleTimeout="180" waitTimeout="0"/> </ConnectionManager> </DatabaseBackend> <RequestManager> <RequestScheduler> <RAIDb-1Scheduler level="passThrough"/> </RequestScheduler> <RequestCache> <MetadataCache/> <ParsingCache/> <ResultCache granularity="table" /> </RequestCache> <LoadBalancer> <RAIDb-1> <WaitForCompletion policy="first"/> <RAIDb-1-LeastPendingRequestsFirst/> </RAIDb-1> </LoadBalancer> <RecoveryLog driver="org.postgresql.Driver" url="jdbc:postgresql:// 192.168.5.1:5432/recovery" login=" " password=" "> <RecoveryLogTable/> <CheckpointTable/> <BackendTable/> <DumpTable/> </RecoveryLog> </RequestManager> </VirtualDatabase> </SEQUOIA> On Sat, Oct 11, 2008 at 12:24 AM, Emmanuel Cecchet <[EMAIL PROTECTED]>wrote: > Hi Bin, > >> I've configuration two distributed sequoia controllers, each connecting to >> a postgresql backend. >> >> The actual Sequoia URL is defined as: >> >> jdbc:sequoia://192.168.5.1 <http://192.168.5.1>, >> 192.168.5.2/test?preferredController=roundRobin < >> http://192.168.5.2/test?preferredController=roundRobin> >> >> when i shutdown any of the controller or any virtualdatabase, the failover >> work fine and Sequoia has no problem of detecting either case of failures. >> However, when i shutdown the backend postgresql, the failover doesn't work, >> just wondering is there any parameter that I can tune for the failover >> timeout or is there anything i can do? >> > If you shutdown the virtual database then the failover should happen. If > you simply disable the backend, the controller is still active and will > forward the requests to the other controller. Note that a controller without > backend still logs all write requests to its recovery log to be able to > recover/restart its backends at any time. If you want to force a failover, > you will have to shutdown the virtual database and not just disable the > backend. > > Thanks for your interest in Sequoia, > Emmanuel > > -- > Emmanuel Cecchet > FTO @ Frog Thinker Open Source Development & Consulting > -- > Web: http://www.frogthinker.org > email: [EMAIL PROTECTED] > Skype: emmanuel_cecchet > > _______________________________________________ > Sequoia mailing list > [email protected] > https://forge.continuent.org/mailman/listinfo/sequoia >
_______________________________________________ Sequoia mailing list [email protected] https://forge.continuent.org/mailman/listinfo/sequoia
