Replicate Multiple HDFS Accounts



Replicate multiple HDFS accounts to one or many databases.

CData Sync for HDFS is a stand-alone application that provides solutions for a variety of replication scenarios such as replicating sandbox and production instances into your database. Both Sync for Windows and Sync for Java include a command-line interface (CLI) that makes it easy to manage multiple HDFS connections. In this article we show how to use the CLI to replicate multiple HDFS accounts.

Configure HDFS Connections

You can save connection and email notification settings in an XML configuration file. To replicate multiple HDFS accounts, use multiple configuration files. Below is an example configuration to replicate HDFS to SQLite:

Windows

<?xml version="1.0" encoding="UTF-8" ?> <CDataSync> <DatabaseType>SQLite</DatabaseType> <DatabaseProvider>System.Data.SQLite</DatabaseProvider> <ConnectionString>Host=sandbox-hdp.hortonworks.com;Port=50070;Path=/user/root;User=root;</ConnectionString> <ReplicateAll>False</ReplicateAll> <NotificationUserName></NotificationUserName> <DatabaseConnectionString>Data Source=C:\my.db</DatabaseConnectionString> <TaskSchedulerStartTime>09:51</TaskSchedulerStartTime> <TaskSchedulerInterval>Never</TaskSchedulerInterval> </CDataSync>

Java

<?xml version="1.0" encoding="UTF-8" ?> <CDataSync> <DatabaseType>SQLite</DatabaseType> <DatabaseProvider>org.sqlite.JDBC</DatabaseProvider> <ConnectionString>Host=sandbox-hdp.hortonworks.com;Port=50070;Path=/user/root;User=root;</ConnectionString> <ReplicateAll>False</ReplicateAll> <NotificationUserName></NotificationUserName> <DatabaseConnectionString>Data Source=C:\my.db</DatabaseConnectionString> </CDataSync>

In order to authenticate, set the following connection properties:

  • Host: Set this value to the host of your HDFS installation.
  • Port: Set this value to the port of your HDFS installation. Default port: 50070

Configure Queries for Each HDFS Instance

Sync enables you to control replication with standard SQL. The REPLICATE statement is a high-level command that caches and maintains a table in your database. You can define any SELECT query supported by the HDFS API. The statement below caches and incrementally updates a table of HDFS data:

REPLICATE Files;

You can specify a file containing the replication queries you want to use to update a particular database. Separate replication statements with semicolons. The following options are useful if you are replicating multiple HDFS accounts into the same database:

You can use a different table prefix in the REPLICATE SELECT statement:

REPLICATE PROD_Files SELECT * FROM Files

Alternatively, you can use a different schema:

REPLICATE PROD.Files SELECT * FROM Files

Run Sync

After you have configured the connection strings and replication queries, you can run Sync with the following command-line options:

Windows

HDFSSync.exe -g MyProductionHDFSConfig.xml -f MyProductionHDFSSync.sql

Java

java -Xbootclasspath/p:c:\sqlitejdbc.jar -jar HDFSSync.jar -g MyProductionHDFSConfig.xml -f MyProductionHDFSSync.sql

Ready to get started?

Learn more or sign up for a free trial:

CData Sync