Tools for Hadoop disaster recovery
This is a set of scripts that interacts with the Ranger REST API to transfer and compare policies of two clusters. Have a look at the README.md in the rangercli subdirectory for more details.
This directory contains a few XML files serving as examples...
-
for creating a cluster entity in Falcon (cluster_example.xml)
-
for creating a mirror process between two clusters (mirror_example.dms)
This is simple sample script that does the following:
- connects to MySQL server
- dumps content of "hive" database on disk (per default in /tmp/ directory)
- changes cluster name
- imports dump to backup MySQL server
- clean up
- Environment variables must be set
CLUSTERNAME, HIVE_USER, HIVE_PASSWORD, MYSQL_HOST, CLUSTERNAME2, HIVE_USER2, HIVE_PASSWORD2, MYSQL_HOST2
- Network connectivity to both hosts on port 3306 must be given.
- User need to be allowed to connect to hive database on both servers from server the script is executed from.
Make sure script has execute permissions.
./hive-schema-copy.sh