You can put IP addresses or host names of the region servers in the regionservers file. If you start HBase using start-hbase.sh on the master, it will use the entries in this file to start each region server for you. Make sure your master can SSH to each region server without a password. The regionservers file isn't used on the region server nodes themselves, but most people keep the config files in sync across the cluster anyway. If you want to stop/start a region server after the cluster is already running, you could use "hbase-daemon.sh stop|start regionserver" from the region server node.
On Thu, Oct 15, 2009 at 9:50 AM, Ananth T. Sarathy < ananth.t.sara...@gmail.com> wrote: > I am little confused about regionserver set up > > after adding > > <property> > <name>hbase.cluster.distributed</name> > <value>true</value> > <description>The mode the cluster will be in. Possible values are > false: standalone and pseudo-distributed setups with managed Zookeeper > true: fully-distributed with unmanaged Zookeeper Quorum (see > hbase-env.sh) > </description> > </property> > > to the hbase-site.xml on the master I still need to edit the regionserver > file. Do i just put the ip addresses of the regionservers i want to add? > > finally, on the instances that I want to run regionservers, do I just use > the start-hbase.sh script? do I need to edit the regionserver file on those > machines as well? Am I missing a page that that explains this more clearly? > I feel kind of stupid asking any of this, but I think i am missing > something. > > Ananth T Sarathy >
