Hi Satish,

To add a component to Ambari, one would do the following:
* Add the component as part of the stack definition (see
https://github.com/apache/ambari/tree/trunk/ambari-server/src/main/resources/stacks/HDP/1.2.0/services
)
* Write Puppet scripts for installing, starting, and testing the component
(see
https://github.com/apache/ambari/tree/trunk/ambari-agent/src/main/puppet/modules
)
* Create an RPM for the component
* Make the RPM is available via yum (or zypper) on the cluster hosts.

Unfortunately we don't have any documentation on this yet.

Yusaku


On Tue, Feb 26, 2013 at 2:01 PM, Satish Abburi <[email protected]>wrote:

>
> we have a columnar DB which is also a distributed component and exploring
> to leverage Ambari to configure this on the Hadoop platform.
>
>
> On Tue, Feb 26, 2013 at 1:55 PM, Mahadev Konar <[email protected]>wrote:
>
>> Hi Satish,
>>  Can you elaborate on the component in question here? Also we are
>> trying to create a more formal doc to be able to add services in the
>> Ambari support stack. Ill probably put something up by this weekend.
>>
>> thanks
>> mahadev
>>
>>
>> On Tue, Feb 26, 2013 at 1:47 PM, Satish Abburi <[email protected]>
>> wrote:
>> >
>> > Any pointers to what API's/extensions we need to use for adding a new
>> > component to the Hadoop stack and have this configure/manage/monitor by
>> > Ambari.
>> >
>> > Our component is packaged as RPM and this needs to be installed on all
>> the
>> > hadoop cluster and configured.
>> >
>> > Thanks,
>> > Satish
>>
>
>

Reply via email to