Its usually safe to default that hbase does things, for better or worse, the way hadoop does (Its probably why there is no doc. on this that is explicity hbase -- would be redundant we feel).
Otherwise, if you'd like to learn more on hbase explicitly, check out the scripts in bin. They are modeled on the hadoop ones. They do stuff like forward ssh environment variables if wanted, and run commands remotely via ssh against machines listed in regionservers file. The commands run remotely are generally those in the bin directory. St.Ack On Thu, Oct 15, 2009 at 11:40 AM, Ananth T. Sarathy < [email protected]> wrote: > thanks... Is there any place I can read up on the inner workings of all > this? I find my ability to google the nitty gritty to be lacking when it > comes to hbase > Ananth T Sarathy > > > On Thu, Oct 15, 2009 at 2:36 PM, Kevin Peterson <[email protected] > >wrote: > > > On Thu, Oct 15, 2009 at 11:05 AM, Ananth T. Sarathy < > > [email protected]> wrote: > > > > > so just by IP address, Hbase will run instances on other servers as > long > > as > > > it can ssh with a password? I mean how does it know where the binaries > > > are? > > > > > > > > I think it does the same thing as Hadoop -- it assumes that the > HBASE_HOME > > is in the same place as on the master node. Look at bin/hbase (it's a > shell > > script) for details. > > >
