Adding new data nodes to Hadoop without rebooting

Usually, I have been wonder how to new data nodes (or recovered nodes) to Hadoop without rebooting. Recently, I found the solution from hadoop core-user mailing list.

The way is very simple as follows:

1. configure conf/slaves and *.xml files on master machine
2. configure conf/master and *.xml files on slave machine
3. run ${HADOOP}/bin/hadoop datanode

If you have to add more than one data node to Hadoop, run the following command (instead of the third command above) on master machine.

${HADOOP}/bin/start-all.sh

Additionally, the way to add a region server to Hbase master without restarting all is similar to that of Hadoop.

1. configure conf/regionservers and *.xml files on master machine
2. configure conf/*.xml files on slave machine
3. run ${HBASE}/bin/hbase regionserver start


OpenOffice 3.0 for Intrepid (Ubuntu 8.10)

OpenOffice 3.0 is released! However, the official repository of Ubuntu 8.10 have not included openoffice 3.0 yet. However, there is a way to enjoy ooo3 in advance. Below link describes the openoffice 3.0 repository for Ubuntu 8.10 Intrepid.

https://launchpad.net/~openoffice-pkgs/+archive

Enjoy OpenOffice 3.0!


Preparing new opensource mirror site for korean users

Recently, I’m preparing the mirroring site for opensource archives. This mirror site aims to provide korean users with opensource archives fastly.

Because the mirroring service consumes so much network bandwidth, we firstly had to be permitted by the office of information technorogy & service. Last week, we finally got the permission from the office.

Above all, we provide the mirror service for Ubuntu, Gentoo Linux, Eclipse, and Apache. Later, if we are asked to mirror other opensource archives, we will do that.

The URL of our mirror site plans to be http://mirror.korea.ac.kr. However, this address is not yet available. I will register this domain to office of infomation technolorogy & service.