The slave replicator will read the THL from the remote master and apply it into Oracle using a standard JDBC connection. The slave replicator needs to know the master hostname, and the datasource type.
Unpack the Tungsten Replicator distribution in staging directory:
tar zxf tungsten-replicator-5.4.tar.gz
Change into the staging directory:
Obtain a copy of the Oracle JDBC driver and copy it into the
cp ojdbc6.jar ./tungsten-replicator/lib/
Install the Slave replicator to read data from the master database and apply it to Oracle:
./tools/tpm install SALES \ --members=host2 \ --master=host1 \ --datasource-type=oracle \ --datasource-oracle-service=ORCL \ --datasource-user=tungsten \ --datasource-password=password \ --install-directory=/opt/continuent \ --svc-applier-filters=dropstatementdata \ --skip-validation-check=InstallerMasterSlaveCheck \ --start-and-report
Once the service has started, the status can be checked and monitored by using the trepctl command.
The description of each of the options is shown below; click the icon to hide this detail:
Click the icon to show a detailed description of each argument.
Specifies the members of the cluster. In this case, the only member is the host into which we are deploying the slave replicator service.
Specify the name of the master replicator that will provide the THL data to be replicated.
Specify the datasource type, in this case Oracle. This configures the replicator to use the Oracle JDBC driver, semantics, and connect to the Oracle database to manager the replication service.
The name of the Oracle service within the Oracle database that
the replicator will be writing data to. For older Oracle
installations, where there is an explicit Oracle SID, use the
command-line option to tpm.
The name of the user created within Oracle to be used for writing data into the Oracle tables.
The password to be used by the Oracle user when writing data.
The directory where Tungsten Replicator will be installed.
Enables a filter that will ensure that statement information is
dropped. When executing statement data that was written from
MySQL, those statements cannot be executed on Oracle, so the
statements are filtered out using the
Skip validation for the MySQL master/slave operation, since that it is irrelevant in a MySQL/Oracle deployment.
Start the service and report the status.
If the installation process fails, check the output of the
for more information about the root cause.
Once the installation has completed, the status of the service should be reported. The service should be online and reading events from the master replicator.