We have configured HUNK with Hadoop 2.6 which is working fine as of now. What we have observed is that when search query is triggered we could see YARN application triggered and it executes successfully.
However, it's using default hadoop configuration (core-default.xml) for the config options which are not specified in Provider, instead of using Site specific configurations (core-site.xml/hdfs/yarn/-site.xml) .
How to make HUNK jobs to use site specific configurations..??
Thanks
You have two options:
1) On the client node / hunk node add the above configurations to the Hadoop default path / classpath.
2) In the Hunk UI - or indexes.conf = add a second provider and make sure you add any Hadoop Client configurations. Just make sure you precedes it with vix. For example, vix.yarn.resourcemanager.ha.rm-ids = rm1,rm2
http://answers.splunk.com/answers/206269/hunk-with-resource-manager-high-availability-is-hu.html