Hi Satish, You can take a look at https://cwiki.apache.org/confluence/display/AMBARI/Ambari+Design which has a design doc attached.
To answer your question, the ambari agent ( python based ) running on the node invokes puppet apply on a generated site.pp ( i.e. puppet is used in standalone mode ). There are plans to start looking at making it easy to add other components to the currently supported stack. If you have any particular requirements that you are trying to address, as a start, filing a jira with more details would help. thanks -- Hitesh On Feb 5, 2013, at 5:11 PM, Satish Abburi wrote: > Thanks. > > Where can i get more info. on how Ambari and Puppet are integrated > architecturally. Also, if we need to add another component to Hadoop stack > and have it installed/configured/managed with Ambari- what is the procedure > to achieve this. Looking for right doc. pointers or any samples for jump > start. > > Thanks, > Satish > > > On Tue, Feb 5, 2013 at 5:06 PM, Siddharth Wagle <[email protected]> > wrote: > Hi Satish, > > You do not need to install puppet. Ambari will do that for you. > > -Sid > > On Tue, Feb 5, 2013 at 4:55 PM, Satish Abburi <[email protected]> wrote: > > Hi, where can i get more insight on this. Do we need to install Puppet before > installing Ambari? > I see some references as Ambari uses Puppet agent... > > Thanks for your help! > > Satish > >
