1 Introduction : 1.1 Pre-Requisites
• the ppas-migrationtoolkit RPM packageBefore installing Migration Toolkit, you must install Java (1.7.0 or higher). For detailed installation instructions for Migration Toolkit, please refer to the EDB Postgres Migration Guide, available at:Before invoking Migration Toolkit to perform a migration, you must download and install a source-specific JDBC driver on your workstation; JDBC drivers are available from the EnterpriseDB website at:After downloading the source-specific driver, move the driver file into the JAVA_HOME/jre/lib/ext directory.Migration Toolkit uses connection information specified in the toolkit.properties file (see Figure 1.1) to connect to the source and target databases. The toolkit.properties file resides in the etc directory, under the edbmtk directory on the workstation from which you will invoke Migration Toolkit.The first three parameters listed in the toolkit.properties file provide connection information for the source database:
• SRC_DB_URL contains a JDBC URL that specifies the type and location of the source database.
• SRC_DB_USER specifies a privileged user of the source database.
• SRC_DB_PASSWORD specifies the password of the source database user.The last three parameters listed in the toolkit.properties file provide connection information for the target (your EDB Ark cluster):
• TARGET_DB_URL contains a JDBC URL of the target database. If you are performing an online migration through an ssh tunnel, specify that the port number of the tunnel in the URL. If you are performing an offline migration, specify the address and port number of your EDB Ark master node when forming the URL.
• TARGET_DB_USER specifies the name of a privileged target database user.
• TARGET_DB_PASSWORD specifies the password of the target database user.When migrating from an Oracle database, SRC_DB_URL takes the form of a JDBC URL. An Oracle URL can take one of two forms:jdbc:oracle:thin:@host_name:port:database_idjdbc:oracle:thin:@//host_name:port/database_id|service_namejdbc identifies the connection protocol.The sub-protocol identifies the migration target as an oracle database.The host_name parameter specifies the name or IP address of the host where the Oracle server is running.port specifies the port number that the Oracle database listener is monitoring.database_id specifies the database SID of the Oracle database.service_name is the name of the Oracle service.For example, the following parameters describe an Oracle source database (named mgmt) that resides on host 192.168.2.88, and is listening on port 1521. Migration Toolkit will attempt to connect using a user name of system, and password of password.Forwarding an unused port on your local workstation through an ssh tunnel to the target cluster allows you to take advantage of ssh compression and encryption during the migration process. To forward a port through an ssh tunnel, use the command:ssh -i ssh_key -C -L local_port:127.0.0.1:target_port user@host_namessh_key specifies the complete path and name of the cluster's ssh key file.-C instructs ssh to use compression.-L instructs ssh to forward the local_port to the remote host and target_port.local_port specifies an otherwise unused port on your local workstation.target_port specifies the port on which the target Postgres server is listening on the EDB Ark cluster. You can find the target_port in the DBPORT column of the Details panel of the Clusters tab in the EDB Ark console (see Figure 1.2).user specifies the name of the connecting user.host_name specifies the name on the EDB Ark cluster with which you wish to connect. You can find the host_name in the DNSNAME column of the Details panel of the Clusters tab in the EDB Ark console (see Figure 1.2).Please note that there is no performance benefit gained by using the load balancing ports during a migration; specify the port number shown in the DBPORT column.
1 Introduction : 1.1 Pre-Requisites