This is a fork of the Roboearth Hadoop Database.
- Clone this repository
- Install [Hadoop] (https://hadoop.apache.org/docs/r1.2.1/single_node_setup.html)
- Install [Hbase] (http://hbase.apache.org/book/quickstart.html)
- Install [HDFS-Fuse] (https://github.com/IDSCETHZurich/roboearthDB/wiki/Install-hdfs-fuse) on hadoop
- Install Apache Tomcat
- Install [Sesame] (http://openrdf.callimachus.net/sesame/2.7/docs/users.docbook?view) on Tomcat
- Make sure to create a user for login for tomcat server apps
- Now login to tomcat and browse to sesame-workbench
- Add a table called 'elements'
- Correct variables in roboearth.py, roboearth.sh and stop-roboearth.sh, settings.py
- Also correct url for static documents in urls.py (The fuse mountpoint)
- Run createTables.py to create necessary tables in Hbase (You need hbase, hadoop and thrift server running)
- Now, you can start the server by running start-roboearth.sh and stop it by stop-roboearth.sh
- The user database contains a dummy user called 'test' to login with password '123456'
- If you are using the AMI, you should start from point 13 of the previous list of instructions
- If you get "could not connect to 'thrift server ip'" (which is mostly on port 9090), Wait sometime. If you still get the error, try restarting roboearth.
- If it takes ages to stop hbase while running stop-roboearth.sh, try pressing Ctrl+C and try again.
- If you reboot an instance, roboearth may not work on the first attempt due to change in ip. Start roboearth and check (ip of instance):60010 to see if hbase is running. If it's not, restart roboearth and wait for sometime for hbase to start. Keep checking logs at hbase/logs/(master log file).