This topic describes how to prepare for using the hive-scanner tool provided by Cloud Migration Hub (CMH). You can learn about the limits on installing and using the tool.
Terms
tool server
The server on which the hive-scanner tool is installed.
Grant user permissions
To use the hive-scanner tool to survey your object resources in a Hive metastore database, you must provide an account that is granted the read permissions on the metastore database of your Hive cluster. The following sample code provides an example on how to grant permissions to the account:
grant select on ${dbname}.* to ${username}@'%';
flush privileges;
Configure a runtime environment
Tool server selection
The hive-scanner tool must be installed and run on a tool server that can communicate with the metastore database of your Hive cluster.
Tool server specifications and OS requirements
The tool server requires at least 2 CPU cores and 4 GB of memory. No special requirements are posed for the operating system (OS) of the tool server.
Environment dependencies
The tool server relies on a Java 8 runtime environment. You can run the java -version command to view the Java version of the tool server.
Download the installation package of the hive-scanner tool
Download the installation package of the hive-scanner tool. Then, upload the installation package to a jump server and decompress it.
OS | Installation package |
Linux (x86)/Windows (x86) | To obtain the installation package, log on to the CMH console. In the left-side navigation pane, choose Assessment Capacity > Resource Collection. On the page that appears, click the Offline Collection tab. Then, click Download Tool in the Hive section. |
To decompress the installation package, run the tar zxvf hive-scanner-x.x.tar.gz command.
Configure a working directory
You must configure an empty directory as the working directory of the hive-scanner tool. Example:
|-hive-scanner/
|-application.yml
|-hms-data-scan-0.0.1-SNAPSHOT.jar
|-start.sh