[
https://issues.apache.org/jira/browse/AMBARI-6148?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14236276#comment-14236276
]
jun aoki commented on AMBARI-6148:
----------------------------------
Hi [~erinaboyd], I wonder if you are still working on this.
I'd like to add a set of service definition for Spark so that spark can be
installed through Ambari.
Please let me know if you can give me this ticket.
I have started investigating a little bit and found a challenge;
Which stack (PHD, BIGTOP and PHD) to add the spark definition first? HDP went
ahead on trunk with versioned rpm so it should be a preferable option but their
yum repo does not provide spark. BIGTOP does but is not versioned rpm yet.
HDP
{code}
bash-4.1# repoquery
--repofrompath=AA,http://s3.amazonaws.com/dev.hortonworks.com/HDP/centos6/2.x/BUILDS/2.2.1.0-2048/
-qa | grep spark
python-sparklines-0:0.9-2.el6.noarch # spark-core is not included
{code}
BIGTOP
http://bigtop01.cloudera.org:8080/job/Bigtop-trunk-Repository/label=centos6/lastSuccessfulBuild/artifact/repo/RPMS/
> Create Spark as a service in HDP
> --------------------------------
>
> Key: AMBARI-6148
> URL: https://issues.apache.org/jira/browse/AMBARI-6148
> Project: Ambari
> Issue Type: Improvement
> Reporter: Erin A Boyd
> Assignee: Erin A Boyd
>
> Extend the HDP 2.1 stack to include Apache Spark as a selectable service
> deployed with HDFS
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
